JP2007124533A - Image pickup apparatus and picked-up image processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enable to ensure a dynamic range required for image output, to reduce deterioration in image quality, and to easily control optical quantity. <P>SOLUTION: In this image pickup apparatus, an incident light through a lens 1 and adjusted by an iris 2 is spread into two light systems in a spectroscope 21. The input light dispersed into two systems is converted into electric signals in image sensors 3, 20. Processing of generating pixel signals from the two systems of electric signals is executed by sample and hold circuit and AGC circuit 5, 24 to linearity adjusting circuits 22, 27. The signal-processed pixel signals of two systems are mixed by a pixel mixing block 23, an image signal is extracted from the mixed pixel signals by a synchronization circuit 9, and the image signal is stored into a dynamic range by a knee circuit 13. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an image capturing apparatus and a captured image processing method applied to, for example, a camera-integrated recording / reproducing apparatus.

FIG. 15 is a block diagram illustrating a configuration of a conventional imaging device.
In FIG. 15, incident light incident through the lens 1 is supplied to the image sensor 3 through the iris 2 as a subject light amount. Here, the image sensor 3 is driven at a timing generated by the timing generation circuit 4. In the image sensor 3, the supplied incident light is converted into an electrical signal corresponding to the subject light amount.

  The output of the image sensor 3 is supplied to a sample hold circuit (S / H) and an AGC (automatic gain control) circuit 5. The sample hold and AGC output of the sample hold circuit and AGC circuit 5 are supplied to the A / D converter 6 and converted into a digital signal. The digital output of the A / D converter 6 is supplied to the black clamp circuit 7 that clamps the black level of the pixel signal, and the black clamped output is supplied to the gain adjustment circuit 8. The output of the gain adjustment circuit 8 is supplied to the synchronization circuit 9, and the synchronization circuit 9 synchronizes the output timing of each pixel signal so that the R (red) signal, G (green) signal, and B (blue) signal are processed. Converted to Here, the pixel signal is a pixel unit signal that is sequentially processed before the R (red) signal, the G (green) signal, and the B (blue) signal are simultaneously extracted.

  This RGB signal is supplied to a detection circuit 15 that detects a reference value and a signal processing system pliny circuit 10 for controlling AE (automatic exposure), AWB (automatic white balance), and the like. Compressed. The signal whose level has been compressed for preprocessing by the pliny circuit 10 is supplied to a gamma correction circuit (1 / γ) 11 for gamma correction for correcting the characteristics of the output image. The signal after the gamma correction is supplied to the LPF 17 which performs low-pass processing for creating a color signal system, and then the color difference signal (R) is adjusted by adjusting the hue (hue) / gain parameter in the hue / gain matrix 18. -Y / BY) and output as an image signal. Here, the image signal is a signal of an image that is finally output.

  On the other hand, the RGB signals are also supplied to the luminance conversion matrix circuit 12, and the output from the luminance conversion matrix circuit 12 is compressed by the knee circuit 13 so as to match the characteristics of the output image. The signal compressed by the knee circuit 13 is supplied to the LPF 14 and output as the luminance signal Y. Each signal processing from the sample and hold circuit and AGC circuit 5 to the hue (gain) / gain matrix 18 and adjustment of the subject light quantity in the iris 2 and the like are performed by the microcomputer 15 via the control bus 19.

As the image sensor 3 described above, an imaging device using a solid-state imaging device such as a CCD (charge coupled device) has been widely used.
However, in consumer imaging devices, unlike commercial use, cost is reduced and size reduction is required, and a dynamic range for incident light cannot be sufficiently obtained. Therefore, when a bright subject and a dark subject are mixed within the shooting angle of view, the bright subject may be overexposed or the dark subject may be collapsed.

Even if the output dynamic range of the image sensor 3 is large, the sample hold circuit (S / H) 5 and the A / D converter 6 in the subsequent stage similarly wait for a sufficient dynamic range to pass the signal. I had to. However, since it is necessary to reduce the power supply voltage of the device in order to reduce power consumption, it is difficult to cope with this point.
For the reasons described above, consumer devices were far from being expensive in terms of image pickup apparatuses and increased in dynamic range for commercial equipment that has a slightly higher level of consumption power than consumer devices.

  Specific contents that cannot secure the wide dynamic range will be described with reference to FIG. FIG. 16 is a diagram for explaining the operation of a conventional imaging apparatus. FIG. 16 shows only the camera signal processing system in order to explain the operation of the conventional imaging apparatus explained in FIG. FIG. 17 is a diagram illustrating an example of a captured image. A case will be described as an example in which an image of the person P and the sky S as illustrated in FIG. 17 is captured by the imaging apparatus as illustrated in FIG. In FIG. 16, the outputs from the lens 1 to the knee circuit 13 are denoted by S1 to S11, and the relationship between the input and the output corresponding to S1 to S11 is illustrated as the horizontal axis and the vertical axis in the characteristic diagrams shown in FIGS. ing.

FIG. 18 is a characteristic diagram of input / output light from the lens to the iris. FIG. 19 is a characteristic diagram of input / output of the image sensor.
When imaging is performed so that the face of the person P is appropriately exposed, the light input from the lens 1 shown on the horizontal axis passes through the iris 2 to the image sensor 3 as output light shown on the vertical axis, as shown in FIG. Entered. Here, the output light of the person P is within the saturation level Is of the image sensor 3, but the output light of the sky S is higher than the saturation level Is of the image sensor 3. However, since a signal equal to or higher than the saturation level Is of the image sensor 3 is not output from the image sensor 3, as shown in FIG. 19, the signal level of the output light of the sky S equal to or higher than Is becomes a constant value SWT.

FIG. 20 is a diagram illustrating an example in which an empty part of the background is overexposed.
That is, the image output as a result described above is as shown in FIG. 20, and the background sky S portion has a constant value SWT, so that the image is overexposed. Even if the output of the image sensor 3 has a sufficient dynamic range, the same result is obtained if the dynamic range of the sample hold circuit (S / H) 5 and the A / D converter 6 in the subsequent stage is not sufficient.

As a method for suppressing the whiteout, there is another method in which the iris 2 suppresses the input light to the image sensor 3 to the saturation level Is of the image sensor 3 where the image sensor 3 is not saturated. Other methods will be described below.
FIG. 21 is a characteristic diagram of input / output light from another lens to the iris. FIG. 22 is a characteristic diagram of input / output of another image sensor.
In this method, as shown in FIGS. 21 and 22, both the output light from the iris 2 and the output light from the image sensor 3 saturate the signal of the empty portion S by suppressing the empty portion SBK of the saturation level Is of the image sensor 3. It is possible to keep the gradation without doing so. However, in this case, the signal quantity required for the person portion PBK cannot be obtained.

For this reason, as a result, the person portion PBK is blackened to the extent that it cannot be distinguished from the background. FIG. 23 shows an example in which the person portion is blackened.
In order to avoid this, the signal is once increased by the gain adjustment circuit 8 and compressed within the dynamic range D of the output image using the pliny circuit 10, the gamma correction circuit 11, and the knee circuit 13. There is a way.

FIG. 24 is a characteristic diagram of input / output of the gain adjustment circuit. FIG. 25 is a characteristic diagram of input / output from the pliny circuit, the gamma correction circuit, and the knee circuit to image output.
As shown in FIGS. 24 and 25, both the output of the gain adjusting circuit 8 and the final image output can secure the target output level as the dynamic range D as the person portion PNA and the empty portion SNA. However, in this case, the output level is compensated by increasing the gain of the person portion PBK and the empty portion SBK having a small signal amount.

For this reason, the noise of the person portion PNA is also amplified, resulting in an image having a poor S / N. FIG. 26 is a diagram illustrating an example in which the noise of the person portion is amplified.
In view of this, a method has been proposed in which a plurality of images with different exposures are captured during one field period, and only a suitable part of the exposure is synthesized to obtain one image to increase the dynamic range ( Patent Document 1).

Similarly, an optimum exposure is selected from among a plurality of image signals obtained by separating a subject by means of using a plurality of photographed images having different exposure times (see, for example, Patent Document 2) or means capable of varying the light amount ratio. There has also been proposed a method in which only one part is collected and a single image is synthesized (see, for example, Patent Document 3). Furthermore, an image dividing optical system has also been proposed in which an optical path is divided into an upper half and a lower half of an image using an optical path dividing member and an imaging unit is arranged at the position of each partial space image (for example, see Patent Document 4).
Japanese Patent Laid-Open No. 7-95481 JP 7-131799 A JP 2001-136434 A JP-A-9-101476

However, in the method described in Patent Document 1, since a plurality of images must be captured in one field, high-speed driving of the image sensor is required, so that it is not suitable for the required increase in the number of pixels. .
In addition, the method described in Patent Document 2 cannot capture a moving object because a time lag occurs in capturing an image. Further, although a method of changing the exposure amount with an aperture or the like has been disclosed, if a plurality of images with different apertures are synthesized, normal image synthesis could not be performed due to a difference in the degree of background blur.

Furthermore, although the method described in Patent Document 3 can increase the dynamic range, it is disadvantageous in terms of sensitivity because a signal of an inappropriate exposure portion is discarded. In addition, since the light amount control for changing the light amount ratio depending on the subject is complicated, it is not suitable for an imaging device that requires real-time performance such as moving image shooting.
Therefore, any method has a disadvantage that the image quality of the captured image is deteriorated. The technique described in Patent Document 4 can increase the resolution of an image, but cannot increase the dynamic range.

  The present invention has been made in view of the above-described problems, and is suitable for moving image shooting that secures a dynamic range necessary for image output, reduces image quality deterioration, and can easily control the amount of light. The object is to provide an apparatus.

  In order to solve the above problems and achieve the object of the present invention, an image pickup apparatus of the present invention includes an optical system that captures input light from a subject, and a light amount adjustment unit that adjusts the amount of light of the subject incident through the optical system. A spectral unit that splits the input light adjusted by the light amount adjusting unit into a plurality of light, a plurality of imaging units that convert the input light split into a plurality of light by the spectral unit into electrical signals, and a plurality of imaging units. A plurality of first signal processing units for generating pixel signals from a plurality of electrical signals, and a mixing unit for mixing a plurality of pixel signals distributed and signal-processed by the plurality of first signal processing units A second signal processing unit that extracts an image signal that can be output as an output image from the pixel signals mixed by the mixing unit, and the image signal extracted by the second signal processing unit is stored in the dynamic range of the output image. A compression unit for those equipped with.

  Further, the captured image processing method of the present invention includes a step of adjusting a subject light amount incident via an optical system, a step of splitting the input light whose light amount has been adjusted into a plurality, and an input split into the plurality of inputs. A step of converting light into a plurality of electrical signals, a step of performing a first signal processing for generating a plurality of pixel signals from the plurality of converted electrical signals, and a plurality of steps by the first signal processing. A step of mixing a plurality of distributed and signal-processed pixel signals, a step of performing a second signal processing for extracting an image signal that can be output as an output image from the mixed pixel signals, and the second signal Compressing the output image so that the image signal extracted by the processing falls within the dynamic range of the output image.

  According to the above-described image pickup apparatus and picked-up image processing method of the present invention, the input light is split into a plurality of parts by the spectroscopic unit (or the step of splitting), and the plurality of pixel signals distributed and signal processed are mixed. The image signal is mixed within the dynamic range of the output image by the compression unit (or the compression step).

  As a result, for example, regarding an imaging device using an imaging device such as a CCD, the dynamic range of the imaging device itself and the dynamic range of the analog device in the subsequent stage, which are the biggest bottleneck in securing a wide dynamic range, Without using expensive and special parts (such as increasing the power supply voltage and widening the dynamic range) like an image pickup device, and using only an inexpensive image sensor and analog device. The expansion of the range can be realized.

  According to the present invention, the input light is divided into a plurality of parts by the spectroscopic unit, a plurality of pixel signals distributed and signal processed are mixed by the mixing unit, and the image signal is converted by the compression unit to the dynamic range of the output image. By being within, the dynamic range necessary for image output is ensured and the image quality of the captured image can be improved.

Hereinafter, embodiments of the present invention will be described with reference to the drawings as appropriate.
FIG. 1 is a block diagram illustrating a configuration of an imaging apparatus according to an embodiment of the present invention. In FIG. 1, components corresponding to those in FIG.
In FIG. 1, incident light incident through the lens 1 enters the spectroscope 21 through the iris 2 as a subject light amount. The spectroscope 21 splits the incident light in half, and the subject light quantity of the incident light split in half is input to the image sensor A shown in 3 and the image sensor B shown in 20, respectively. The image sensors 3 and 20 are driven at the timing generated by the timing generation circuit 4.

  In the image sensors 3 and 20, the subject light amount of incident light supplied from the spectroscope 21 is converted into two systems of electrical signals. Outputs of the image sensors 3 and 20 are supplied to sample and hold circuits and AGC circuits 5 and 24 of different systems, respectively. Next, the outputs of the two systems of sample and hold circuits and AGC circuits 5 and 24 are supplied to A / D converters 6 and 25 of different systems, respectively, and converted into digital signals. The digital outputs of the two systems of A / D converters 6 and 25 are supplied to black clamp circuits 7 and 26 that clamp the black level of the pixel signals of different systems.

  The outputs whose black level is clamped by the two black clamp circuits 7 and 26 are supplied to separate linearity adjustment circuits 22 and 27, respectively. A correction value storage area 24 for correcting the linearity using the two systems of linearity adjustment circuits 22 and 27 and storing the two systems of adjustment data is connected to the microcomputer 15. The signal processing from the sample hold circuit and AGC circuits 5 and 24 to the hue (gain) / gain matrix 18 and the adjustment of the subject light quantity in the iris 2 and the like are performed by the microcomputer 15 via the control bus 19.

  Next, the two outputs of the linearity adjustment circuits 22 and 27 are supplied to the pixel mixing block 23. In this pixel mixing block 23, the respective pixel signals of the two systems are mixed. At this time, it is assumed that the bit length of the pixel mixing block 23 has a bit length sufficient to allow a level obtained by adding the maximum levels of the two systems of signals A and B. Further, the spectroscope 21 is split into two systems so that the light quantity is within the dynamic range of signal processing in the two systems of sample and hold circuits and AGC circuits 5 and 24 and A / D converters 6 and 25.

  Next, the output obtained by pixel mixing in the pixel mixing block 23 is supplied to the gain adjustment circuit 8. The output of the gain adjustment circuit 8 is supplied to the synchronization circuit 9, and the synchronization circuit 9 converts the output timing of each pixel signal into the R signal, the G signal, and the B signal by performing the synchronization processing. Here, the pixel signal is a pixel unit signal that is sequentially processed before the R (red) signal, the G (green) signal, and the B (blue) signal are simultaneously extracted. This RGB signal is supplied to a detection circuit 15 that detects a reference value for control of AE, AWB, and the like, and a Pliny circuit 10 that performs signal level compression processing for pre-processing of a subsequent signal processing system. The signal whose level has been compressed for preprocessing by the pliny circuit is supplied to the gamma correction circuit 11 and subjected to gamma correction for correcting the characteristics of the output image.

The signal after the gamma correction is supplied to the LPF 17 that performs low-pass processing for creating a color signal system represented by a color difference signal. The signal after the low-pass processing is converted into a color difference signal (RY / BY) by adjusting the parameters of hue (hue) / gain in the hue (gain) / gain matrix 18 and output as an image signal. Here, the image signal is a signal of an image that is finally output.
On the other hand, the RGB signal is also supplied to the luminance conversion matrix circuit 12, and this output is supplied to the LPF 14 after the signal level is compressed by the knee circuit 13, and the signal after the low-pass processing is the luminance signal Y. Is output as

FIG. 2 is a diagram for explaining the operation of the imaging apparatus.
An actual operation in the imaging apparatus shown in FIG. 1 will be described with reference to FIG. This figure represents only the camera signal processing system in the imaging apparatus of the present embodiment described in FIG. A case where an image of the person P and the sky S is captured using the imaging apparatus as shown in FIG. 2 as in FIG. 17 shown as the prior art will be described as an example. In FIG. 2, the outputs from the lens 1 to the knee circuit 13 are S1 to S11, and the respective inputs and outputs are shown corresponding to the signals on the vertical axis and the horizontal axis in the characteristic diagrams of FIGS. Note that the outputs of the two systems of the sample hold circuit and AGC circuits 5 and 24 to the linearity adjustment circuits 22 and 27 are channel A and channel B.

FIG. 3 is a characteristic diagram of input / output light from the lens 1 to the iris 2.
As shown in FIG. 3, the output light from the iris 2 shown on the vertical axis is input to the spectroscope 21 through the iris 2 with respect to the input light of the lens 1 from the subject shown on the horizontal axis. Here, the output light of the person P is within the saturation level Is of the image sensor 3, but the output light of the sky S is higher than the saturation level Is of the image sensor 3.

FIG. 4 is a characteristic diagram of the input / output light of the spectroscope. FIG. 4A is channel A, and FIG.
As shown in FIGS. 4A and 4B, the spectroscope 21 splits the input light into the channel A and the channel B in half and outputs them. As shown in FIGS. 4A and 4B, the light amounts of the channels A and B are halved, so that the output light of the persons PA and PB and the sky SA and PB are all at the saturation level Is of the image sensor. Can fit inside.

5A and 5B are graphs showing input / output characteristics of the image sensor. FIG. 5A shows channel A, and FIG.
As described above, since the amount of light in each of the channels A and B is halved, as shown in FIGS. 5A and 5B, the output levels of the persons PA and PB and the sky SA and PB from the image sensor 3 are as follows. It is possible to extract all signal levels in a desired region so as to be within the input dynamic range Is of the sample hold circuit (S / H) and the AGC circuits 5 and 24.

  Here, in consideration of the maximum output Is of the image sensor 3, signal processing before pixel mixing processing such as the sample-and-hold circuit (S / H) and the AGC circuits 5 and 24, A / D converters 6 and 25, etc. in the subsequent stage If the system is selected, that is, if each of the channels A and B is set, the output from the image sensor 3 is input to the input stage of the pixel mixing block 23 while maintaining the dynamic range as it is. Will be.

FIG. 6 is a characteristic diagram of input / output of the pixel mixture block.
The person PA and PB and sky SA and PB output signals of channel A and channel B input to the pixel mixing block 23 are added in the pixel mixing block 23, and the output of the person P and sky S as shown in FIG. Get.

  Here, the input dynamic range of the signal processing system that is the output from the pixel mixing block 23 with respect to the dynamic ranges Is of the persons PA and PB and the empty SA and PB of the channels A and B input to the pixel mixing block 23. GMo is doubled. In the signal processing system after the pixel mixing block 23, the dynamic range GMo secures a bit length sufficient to pass a signal amount twice as large as Is.

  The output from the pixel mixing block 23 is input to the pliny circuit 10. In the pliny circuit 10, when the bit length is insufficient in the subsequent stage, the high frequency portion is compressed in advance here. If the bit length of the subsequent signal processing is sufficient, the signal may be input to the gamma correction circuit 11 without being compressed here. Here, it is assumed that the bit length of the subsequent stage is sufficient, and the compression process by the pliny circuit 10 is omitted.

FIG. 7 is a characteristic diagram of input / output of the γ circuit (gamma correction circuit).
The output of the pixel mixing block 23 is directly input to the gamma correction circuit 11, and an output after gamma correction as shown in FIG. 7 is obtained. Here, the output maximum value γo of the gamma correction circuit 11 is larger than the dynamic range So of the final output image, and the signal in the empty S portion is cut off as it is.

FIG. 8 is a characteristic diagram of input / output of the knee circuit. FIG. 9 is a characteristic diagram of input / output from the knee circuit to image output.
For this reason, the knee circuit 13 compresses the knee circuit output so as to be within the dynamic range So of the final output image by the knee characteristic as shown in FIG.
As a result, an image output in which the signal of the empty S ′ portion as shown in FIG. 9 is finally compressed so as to be within the dynamic range So of the final output image is obtained.

  If this method is used, the required signal level is obtained by mixing the pixels of channel A and channel B, so that the reduction in S / N can be suppressed and the dynamic range can be doubled. In addition, although it is very difficult to expand the dynamic range in an analog device due to problems such as downsizing and power consumption, the method of this embodiment does not require increase in power consumption, and relatively easy digital signal processing is possible. This can be realized only by securing the bit length.

  In the example described above, the case where the characteristics of the spectroscope 21, the image sensors 3 and 20, the sample hold circuit and AGC circuits 5 and 24, and the A / D converters 6 and 25 are ideal has been described. The characteristic variation of the spectroscope 21 and variations in linearity indicating the linearity of sensitivity and processing characteristics due to the image sensors 3 and 20, the sample hold circuit and AGC circuits 5 and 24, and the A / D converters 6 and 25 are generated. To do.

In the following, correction for variations in nearness will be described.
FIG. 10 is a diagram illustrating variation in characteristics of the spectrometer. FIG. 11 is a diagram illustrating variations in image sensors.
As shown in FIG. 10, the output from the spectroscope 21 has a level difference due to the characteristic variation of the spectroscope 21 from the solid line to the dotted line between the channel A and the channel B. When the output from the spectroscope 21 is input to the image sensors 3 and 20, the output of the image sensors 3 and 20 corresponds to the sensitivity difference between the solid line and the dotted line by the image sensors 3 and 20 and the variation in linearity. Therefore, the variation increases as shown in FIG.

  After adjusting the level difference between the output levels of the image sensors 3 and 20 and the difference between the black levels by the sample hold (S / H) and the AGC 5, the black clamp circuits 7 and 26 adjust the starting point to correct the difference between them. Then, the level difference between channel A and channel B is corrected.

FIG. 12 is a diagram illustrating level difference correction by the black clamp circuit from the sample hold (S / H).
Due to this correction, the level difference due to the black clamp circuit from the sample hold (S / H) in the channel A and channel B indicated by the thin solid line and the dotted line in FIG. 12 is changed as indicated by the thick solid line and the dotted line. The variation between channel A and channel B is improved.
However, there is still a difference in linearity between channel A and channel B, and this leads to deterioration of gradation characteristics and S / N. Therefore, the linearity adjustment circuits 22 and 27 correct the linearity of channel A and channel B. To do.

FIG. 13 is a diagram illustrating a difference between an ideal straight line and actual input / output.
As shown in FIG. 13, in the linearity adjustment circuits 22 and 27, the difference between the ideal straight line and the actual input / output signal is measured in advance so that an ideal input / output characteristic is obtained, and the microcomputer 15 stores the correction value storage area. The correction value is stored in 24.

  Accordingly, the linearity adjustment circuits 22 and 27 perform correction by subtracting a difference from the correction value based on the correction value, and an output with linearity between the channel A and the channel B is obtained. FIG. 14 is a diagram illustrating a result of correcting input / output by the linearity adjustment circuit.

If the difference in linearity between channel A and channel B is not so large, only the linearity on one side may be corrected and matched with the other linearity.
Based on this, pixel mixing and signal level compression by the signal processing system as described above minimizes S / N deterioration and reduces the dynamic range without losing sensitivity. It becomes possible to do.

  According to the present embodiment, it is possible to A / D convert the output of an inexpensive image sensor and the image sensor into a digital signal without using expensive parts as in a commercial imaging apparatus. The analog signal processing part of the system can realize a wide dynamic range without using a special part that can take a large dynamic range (for example, a high power supply voltage level). It is possible to provide an image pickup apparatus that is free from problems such as deterioration of the image quality. Since there is no need to increase the power supply voltage, it also leads to a reduction in power supply, which leads to a total power reduction.

  In addition, in order to suppress the variation between the two signal systems, device pairing or high-spec parts are usually required. Since the correction processing regarding the variation of the image sensor and the variation of the analog device is also performed at the same time, it is possible to reduce the above-described restriction. In the above-described example, the light and the signal are distributed to the two systems by the spectroscope. However, not only the two systems but also the three systems and the four systems are distributed in order to be within the dynamic range of the signal processing system. May be.

  It goes without saying that the configuration can be changed as appropriate within the scope of the present invention, not limited to the embodiment described above.

It is a block diagram which shows the structure of the imaging device by implementation of this invention. It is a figure explaining operation | movement of an imaging device. It is a characteristic view of the input / output light from a lens to an iris. FIG. 4A is a channel A and FIG. 4B is a channel B. FIG. FIG. 5A is a channel A and FIG. 5B is a channel B. FIG. It is an input-output characteristic figure of a pixel mixing block. It is an input-output characteristic figure of a gamma circuit. It is an input / output characteristic diagram of the knee circuit. It is a characteristic diagram of input / output from a knee circuit to image output. It is a figure which shows the dispersion | variation in the characteristic of a spectrometer. It is a figure which shows the variation of an image sensor. It is a figure which shows correction | amendment of the level difference by a black clamp circuit from sample hold (S / H). It is a figure which shows the difference of an ideal straight line and actual input / output. It is a figure which shows the correction | amendment of the input / output by a linearity adjustment circuit. It is a block diagram which shows the structure of the conventional imaging device. It is a figure explaining operation | movement of the conventional imaging device. It is a figure which shows the example of a captured image. It is a characteristic view of the input / output light from a lens to an iris. It is an input-output characteristic figure of an image sensor. It is a figure which shows the example in which the sky part of a background skips white. FIG. 6 is a characteristic diagram of input / output light from another lens to an iris. It is a characteristic view of the input / output of another image sensor. It is a figure which shows the example in which a person part is crushed black. It is an input-output characteristic figure of a gain adjustment circuit. It is a characteristic diagram of input / output from a pliny circuit, a γ circuit, and a knee circuit to image output. It is a figure which shows the example in which the noise of a person part is amplified.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Lens, 2 ... Iris, 3 ... Image sensor, 4 ... Timing generation circuit, 5 ... Sample hold circuit and AGC circuit, 6 ... A / D converter, 7 ... Black clamp circuit, 8 ... Gain adjustment circuit, 9 ... Synchronizing circuit, 10 ... Pliny circuit, 11 ... Gamma correction circuit, 12 ... Luminance conversion matrix, 13 ... Knee circuit, 14 ... LPF, 15 ... Microcomputer, 16 ... Detection circuit, 17 ... LPF, 18 ... Hue (color) / Gain matrix, 19 ... control bus, 20 ... image sensor, 21 ... spectrometer, 22 ... linearity adjustment circuit, 23 ... pixel mixing block, 24 ... sample hold circuit and AGC circuit, 25 ... A / D converter, 26 ... Black clamp circuit, 27 ... linearity adjustment circuit, 28 ... correction value storage area

Claims (6)

An image imaging device that performs signal processing on an image signal obtained by imaging a subject and outputs an output image with a predetermined dynamic range,
An optical system that captures input light from the subject,
A light amount adjustment unit for adjusting the amount of light of the subject incident through the optical system;
A spectroscopic unit that splits the input light adjusted by the light amount adjusting unit into a plurality of parts,
A plurality of imaging units each converting the input light split into a plurality of light by the spectroscopic unit into electrical signals;
A plurality of first signal processing units for generating pixel signals from a plurality of electrical signals converted by the plurality of imaging units;
A mixing unit that mixes a plurality of pixel signals distributed and signal-processed by the plurality of first signal processing units;
A second signal processing unit that extracts an image signal that can be output as an output image from the pixel signals mixed by the mixing unit;
A compression unit for storing the image signal extracted by the second signal processing unit within a dynamic range of an output image;
An image pickup apparatus comprising:   2. The image pickup according to claim 1, wherein the plurality of first signal processing units include a gain adjustment unit for correcting variation of a plurality of electric signals output from the plurality of image pickup units. apparatus.   The plurality of first signal processing units includes a linearity adjustment unit for adjusting linearity of a plurality of generated pixel signals, an adjustment level storage unit that stores an adjustment level of the linearity adjustment unit, and the gain And a control unit that controls the adjustment unit and / or the linearity adjustment unit to perform correction for correcting variations in the plurality of electrical signals output from the plurality of imaging units. 2. The image pickup apparatus according to 2.   2. The image capturing apparatus according to claim 1, wherein the mixing unit is provided at a subsequent stage of a black clamp unit that clamps black levels of a plurality of pixel signals provided in the plurality of first signal processing units.   2. The image capturing apparatus according to claim 1, wherein the spectroscopic section is split into a plurality of light beams so as to have a light amount within a dynamic range of signal processing in the plurality of first signal processing sections. A captured image processing method for performing signal processing on an image signal obtained by imaging a subject and outputting an output image with a predetermined dynamic range,
Adjusting the amount of subject light incident through the optical system;
Spectroscopically splitting the input light whose light amount has been adjusted,
Converting each of the input light split into a plurality of the signals into a plurality of electrical signals;
Performing first signal processing for generating a plurality of pixel signals from the plurality of converted electrical signals;
Mixing a plurality of pixel signals distributed and signal-processed by the first signal processing;
Performing a second signal processing for extracting an image signal that can be output as an output image from the mixed pixel signal;
Compressing the image signal so that the image signal extracted by the second signal processing falls within the dynamic range of the output image;
A captured image processing method comprising:

Images