JP2003333432A - Image pickup device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enable an image pickup device that has an image pickup section having a plurality of image pickup regions and the outputs from the regions to have the relative accuracy of a plurality of outputs. <P>SOLUTION: This image pickup device has the image pickup section (101) which picks up the image of an object by dividing the image into a plurality of image pickup regions and outputs signals from the image pickup regions, signal level distribution detecting means (205 and 206) which respectively detect the signal level distributions of the plurality of output signals outputted from the image pickup regions, and divided signal correcting means (130 and 131) which corrects the plurality of output signals based on the detected results of the detecting means (205 and 206). The detecting means (205 and 206) detect the errors of the plurality of output signals with high accuracy and the correcting means (130 and 131) correct the errors between the plurality of output signals. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image pickup apparatus having an image pickup section for dividing a subject image into a plurality of image pickup areas, picking up an image, and outputting image data from each of the plurality of image pickup areas.

[0002]

2. Description of the Related Art Japanese Patent Application Laid-Open No. 2001-94886 discloses a conventional image pickup apparatus having an image pickup section which divides a subject image into a plurality of image pickup areas, and outputs image data from each of the plurality of image pickup areas. There is one disclosed in the publication. In this conventional imaging device, for a plurality of output signals respectively output from a plurality of output terminals, a region having high correlation and a region having low correlation between the plurality of image data are determined, and the correlation is determined based on the determination. It is characterized in that it has a correction means for correcting a plurality of image data by using image data between regions having high property. In this case, the average value, the maximum value, the minimum value, and the variance value of a plurality of image data are used for the above-mentioned determination of the correlation, and a gain adjusting circuit and an offset adjusting circuit are provided for correcting the image data. .

[0003]

However, in the conventional image pickup apparatus, since correction is performed by linear processing such as gain adjustment and offset adjustment as correction means, accurate correction cannot be performed when nonlinear processing is included. However, since the correction data is calculated using only the image data, there is a problem in that the correction accuracy changes depending on the subject.

Therefore, in the present invention, in the case of an image pickup apparatus having such an image pickup unit having a plurality of outputs, attention has been paid to a specific method of maintaining the relative accuracy of the plurality of outputs.

An object of the present invention is to correct unbalance of image data output from a plurality of output terminals of an image pickup section.

[0006]

In order to solve this problem, the image pickup apparatus of the present invention is constructed as follows.

An image pickup apparatus of the first invention of the present application divides a subject image into a plurality of image pickup areas, picks up an image, and outputs a signal from each of the plurality of image pickup areas, and each of the plurality of image pickup areas. It has a signal level distribution detecting means for a plurality of output signals output from the device and a divided signal correcting means for correcting the plurality of output signals based on the detection result of the signal level distribution detecting means.

Here, the signal level distribution detection means detects the signal level distribution (histogram or the like) of each output signal from the plurality of image pickup areas in the image pickup section, and passes the detection result to the divided signal correction means. The divided signal correction means corrects each output signal from the plurality of imaging regions based on the received signal level distribution detection result. Since the plurality of output signals are corrected based on the signal level distribution, the imbalance between the plurality of output signals can be corrected with high accuracy even when the subject is photographed with variations in nonlinear elements such as changes in contrast.

According to a preferred aspect of the above, the signal level distribution detecting means does not only detect the signal level distribution of the plurality of output signals output from each of the plurality of imaging regions, but also the entire plurality of output signals. Is configured to detect the signal level distribution of

In this way, the area for detecting the signal level distribution serving as the correction reference becomes wide, and the correction of each of a plurality of output signals makes the processing contents of each signal common. A difference between a plurality of output signals due to a difference in signal processing does not occur, and more accurate correction can be performed.

Further, in the above, another preferred embodiment is
The signal level distribution detecting means includes detection of a signal level of a black area in a plurality of output signals output from each of the plurality of imaging areas.

With this configuration, the offset correction is also performed based on the detection of the signal level of the black area, so that the correction can be performed with higher accuracy.

Further, the image pickup apparatus of the second invention of the present application divides a subject image into a plurality of image pickup areas, picks up an image, and outputs a signal from each of the plurality of image pickup areas, and the plurality of image pickup areas. Reliability detection for determining the reliability of the first detection means for the plurality of output signals output from each of the first detection means and the second detection means for the plurality of output signals based on the detection result of the first detection means. And a divided signal correction means for correcting the plurality of output signals based on the detection result of the second detection means which the reliability detection means determines to have reliability.

In correcting the difference between a plurality of output signals, a signal in a saturated region, a signal in a boundary region between different objects, etc. are not suitable as reference signals for correction. That is, the reliability is poor. Therefore, the reliability is determined by whether or not the signal levels of the plurality of output signals are within a certain range, and whether the signal level differences between the plurality of output signals are within a certain range. Then, if the signal in the unreliable region is removed and the detection signal in the case of the reliability is used to perform the correction between the plurality of output signals, the correction can be performed with higher accuracy.

In a preferred mode in the above, the detection result of the second detecting means, which is determined by the reliability detecting means to be reliable, is the black area in the plurality of output signals output from each of the plurality of imaging areas. Is the detection result of the signal level of.

With this configuration, in addition to the reliability judgment,
By performing the offset correction together based on the detection of the signal level of the black area, it is possible to further correct the plurality of output signals.

Further, in the above, another preferable embodiment is
The second reliability determination means determines that there is reliability.
The detection result of the detection means is the detection result of the second detection means immediately before the reliability detection means determines that there is reliability.

As a factor for determining the reliability, there are a difference in processing of paths of a plurality of output signals and a difference between respective subjects.

If the result of the reliability judgment is due to the difference between the subjects, there is a high possibility that the correction will be erroneous. The detection data at this time is not used, and the correction is performed based on what is determined to be reliable immediately before. This makes it possible to perform highly reliable correction without erroneous correction.

In the above, the first detecting means and the second detecting means may be any combination of the signal level distribution detecting means, the signal average value detecting means and the signal peak detecting means.

In the above, the first detecting means may be the signal level distribution detecting means, and the second detecting means may be the signal average value detecting means or the signal peak detecting means. .

Further, the image pickup apparatus according to the third invention of the present application has an area for storing a predetermined charge, divides a subject image into a plurality of image pickup areas, picks up an image, and outputs a signal from each of the plurality of image pickup areas. An image pickup unit, a signal level distribution detection unit for a plurality of output signals output from each of the plurality of image pickup regions, and a division signal correction for correcting the plurality of output signals based on the detection result of the signal level distribution detection unit And means.

The operation of this configuration is as follows.
That is, the area in which the predetermined charge is stored in the image pickup unit can be referred to as a pilot charge area. However, the pilot charge area in the image pickup section is a different area from the image detection area in which the subject appears, and The multiple output signals from the charge region are unaffected by the subject.
Since the plurality of output signals from the image detection area are corrected based on the signals between the plurality of output signals from the pilot charge area that are not affected by the subject, more accurate correction can be performed.

In a preferred aspect of the above, the signal level distribution detecting means detects the signal level distribution of a plurality of output signals output from a region (pilot charge region) storing a predetermined charge in a plurality of image pickup regions. It is said that.

In the above, the preferable mode of the signal level distribution detecting means is configured to detect the signal level distribution of the plurality of output signals output from the plurality of imaging regions corresponding to the subject image. Is. This enables more accurate correction.

Further, the signal level distribution detecting means is
It is further preferable that it is configured to include detection of the signal levels of the plurality of output signals output from the optical black level regions in the plurality of imaging regions. The offset correction and the gain correction can be corrected with higher accuracy.

Then, it is preferable that the predetermined charges are injected from the outside of the image pickup device. Furthermore, it is preferable that the predetermined charge is injected from the outside of the image pickup device at a charge level in a desired range. By adjusting the level of the injected charges, it is possible to perform more accurate correction.

In each of the above-mentioned solving means, as a preferable aspect from another viewpoint, the above-mentioned divided signal correcting means corrects a plurality of output signals based on the detection result of the signal level distribution detecting means. It is preferable to include a correction data calculation means for calculating

Still another preferred aspect is correction data calculation means for calculating a correction amount for the divided signal correction means to correct a plurality of output signals based on the detection result of the signal level distribution detection means, It is to have at least one of an offset correction means for performing offset correction on the plurality of output signals based on the output of the correction data calculation means and a histogram correction means for performing histogram correction.

By each of these correction means, even when the subject is photographed with variations in non-linear elements such as changes in contrast, it is possible to favorably correct the imbalance between the plurality of output signals.

Further, it is preferable that the histogram correction means is configured to make the appearance frequencies of the respective gradations equal to each other for the plurality of output signals. Thereby, the non-linear imbalance between the plurality of output signals can be corrected with higher accuracy.

Further, the image pickup apparatus of the fourth invention of the present application has an area for storing a predetermined charge, divides a subject image into a plurality of image pickup areas, picks up an image, and outputs a signal from each of the plurality of image pickup areas. An image pickup section, a charge signal correcting means for correcting a plurality of output signals output from the area for storing the predetermined charges, a signal level detecting means for the signal corrected by the charge signal correcting means, and a signal level detecting means. And a divided signal correcting means for correcting the plurality of output signals based on the detection result of 1.

The operation of this configuration is as follows.
That is, if there are variations in the pilot charges in the pilot charge region, the accuracy of correction for a plurality of output signals becomes low. Therefore, by adding a charge signal correction means to correct the variation of the pilot charge, the accuracy of the correction using the pilot charge can be made higher.

In the above-mentioned preferred mode, the charge signal correction means has a pixel change detection circuit for detecting the amount of change in pixel units of the plurality of output signals output from the region storing the predetermined charges. .

The operation of this configuration is as follows.
That is, the level of the injected pilot charge changes in pixel units as it advances. The correction using the non-uniform pilot charge reduces the correction accuracy. Therefore, a pixel variation detection circuit is added to correct the non-uniformity of the pilot charges by detecting the variation amount in pixel units, and then a plurality of output signals are corrected. As a result, the accuracy of correction can be made even higher.

Further, in the above, another preferable embodiment is
The charge signal correction means includes a histogram detection circuit that detects a variation amount of the signal level distribution of the plurality of output signals output from the region that stores the predetermined charges. Similar to the above, the accuracy of correcting a plurality of output signals can be made even higher.

Further, in the above, another preferred embodiment is
The charge signal correction means has an averaging correction circuit based on the amount of fluctuation detected from the plurality of output signals output from the area storing the predetermined charges.

In addition, in the above, another preferred embodiment is
The charge signal correction means has a gain correction circuit based on the amount of fluctuation detected from the plurality of output signals output from the region storing the predetermined charges.

By correcting the non-uniformity of the pilot charges and then performing the average value correction and the gain correction, the accuracy of the correction of the plurality of output signals can be further increased.

Finally, the image synthesizing means is provided in each of the solving means. That is, it is an image synthesizing means for synthesizing the plurality of output signals corrected by the divided signal correcting means.

[0041]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of an image pickup apparatus of the present invention will be described below with reference to the drawings.

(First Embodiment) FIG. 1 is a block diagram showing an example of the hardware configuration of an image pickup apparatus according to the first embodiment of the present invention. In FIG. 1, 101 is a CCD which is a plurality of (here, two) image pickup areas and an image pickup unit which independently outputs (CH1 and CH2) signals of the plurality of image pickup areas.
And the like, 102 is a timing signal creation circuit that creates timing signals necessary for driving and signal processing of the imaging device 101, 103 and 104 are analog signal processing circuits that perform noise removal and gain control, and 105 and 106 are analog An analog-digital conversion circuit (hereinafter referred to as an A / D conversion circuit) for converting a signal into a digital signal, 130, 1
Reference numeral 31 is a division signal correction circuit which is a correction means for a difference between channels of two signals, 111 is an image synthesizing circuit for synthesizing a plurality of video signals, 112 is a camera signal processing circuit, and 113 is an inter-channel detecting a difference between a plurality of signals. A detection circuit, 114 is an overall control circuit for controlling the above circuits, and 115 is a mode setting circuit.

The operation of the image pickup apparatus configured as described above will be described below.

In FIG. 1, an image pickup device 101 divides a subject image into a plurality of image pickup regions by a drive signal from a timing signal generation circuit 102, picks up an image, and outputs two signals (CH1 and CH2) from each of the plurality of image pickup regions. ) Is output. The data output from the right half surface of the image sensor 101 is input to the analog signal processing circuit 104 via the CH1 output, and is included in the output of the CCD or the like by performing a known method such as correlated double sampling. The AGC circuit for removing the reset noise and the like and operating the AGC circuit for amplifying the output to a predetermined signal level is operated. By inputting the output after the AGC to the A / D conversion circuit 106, it is converted into a digital signal and an output of AD-CH1 is obtained.

Similarly, the data output from the left half surface of the image pickup device 101 is input to the analog signal processing circuit 103 via the CH2 output, and the same method such as correlated double sampling is performed here, so that the CCD or the like can be obtained. The reset noise and the like included in the output of 1 are removed, and the AGC circuit for amplifying the output to a predetermined signal level is activated. This AG
By inputting the output after C to the A / D conversion circuit 105,
It is converted to a digital signal and an output of AD-CH2 is obtained.

Next, the output signals of AD-CH1 and AD-CH2 are simultaneously input to the inter-channel detection circuit 113 which is a detection means for detecting the difference between the channels of both output signals, and the inter-channel detection circuit 113 is detected. The detection data detected in (4) is input to the overall control circuit 114.

Further, the output signals of AD-CH1 and AD-CH2 are input to divided signal correction circuits 130 and 131 which are correction means for correcting the difference between the channels of both output signals, and these divided signal correction circuits are inputted. At 130 and 131, the predetermined correction control signals CONT1 and CONT calculated and set by the overall control circuit 114, which is the other input signal.
2 is used to correct the error between the two signals.

In this way, the two output signals are detected by the inter-channel detection circuit 113, and the overall control circuit 1
Correction control signals CONT1 and CONT2 calculated in 14
The difference between channels is corrected by the divided signal correction circuits 130 and 131 which are correction means for performing correction based on

In this way, the image data output in which the difference between the channels generated between the two outputs is corrected is converted into one image data by the image synthesizing circuit 111 (the left and right outputs are made into one output), and the next stage is performed. The camera signal processing circuit 112 performs predetermined signal processing (color interpolation processing, γ conversion, etc.).

Next, a specific configuration of the inter-channel detection circuit 113 will be described using the configuration example shown in FIG.

In FIG. 2, 201 is a detection area setting circuit for setting a detection area under the control of the overall control circuit 114 shown in FIG. 1, and 202 is a reliability area for detecting a reliable area for an input signal. Detecting circuits, 203 and 204 are average value detecting circuits for detecting the average value of the input signal in a set range, 205 and 206 are signal level distribution detecting means, which are histograms (signal level distribution) of the input signal in a certain set range. ) Is a histogram detection circuit, and 230 and 231 are input signal peak detection circuits in a set range, and the operation of the inter-channel detection circuit 113 thus configured will be described below.

In FIG. 2, first, the detection area setting circuit 2
The area for detecting the difference between channels is set by 01. The detection area setting circuit 201 determines the effective range for the output signal from the image sensor 101 with reference to the VD / HD signal from the timing signal generation circuit 102 shown in FIG. Set. Further, the reliability area detection circuit 202 is AD-CH.
1, reliability of AD-CH2 is evaluated, a reliable area is detected, and information on the reliable area is set in each detection circuit.

Next, in the average value detection circuits 203 and 204 in each channel, for example, using the reliable area indication signal detected by the reliability area detection circuit 202 in the detection area set by the detection area setting circuit 201, for example, The average value of the signals corresponding to the black level and the average value of the entire signals are detected. Further, the histogram detection circuits 205 and 206 detect the signal level distribution using the reliable area indication signal detected by the reliability area detection circuit 202 in the detection area set by the detection area setting circuit 201. Further, the peak detection circuits 230 and 231 detect the peak level of the signal using the reliable area indication signal detected by the reliability area detection circuit 202 in the detection area set by the detection area setting circuit 201. By using these detected data, correction data of the difference between channels is obtained in the overall control circuit 114 shown in FIG.

The correction data for the difference between the channels is shown in FIG.
The difference between the channels of the divided signals is corrected by the divided signal correction circuits 130 and 131 shown in FIG.

Next, the specific configuration of the divided signal correction circuits 130 and 131 will be described using the configuration example shown in FIG.

In FIG. 3, 301 and 302 are generated by a difference in addition / subtraction processing between channels using predetermined offset correction control signals OF1 and OF2 calculated and set by the overall control circuit 114 which is the other input signal. It is an offset correction circuit that performs addition / subtraction correction on an input signal to correct an offset error. Also, 30
3, 304 is the other input signal, which is a predetermined gain correction control signal GA calculated and set by the overall control circuit 114.
1 is a gain correction circuit that uses 1 and GA2 to perform gain correction on an input signal to correct a gain error caused by a difference in multiplication processing between channels. Also,
Reference numerals 305 and 306 are histogram correction circuits that correct the error in the distribution of the signal level between channels by using the predetermined histogram correction control signals HG1 and HG2 that are calculated and set by the overall control circuit 114 that is the other input signal. . Here, as for the correction of the difference between channels, any one or two of the offset correction, the gain correction, and the histogram correction may be performed according to circumstances.

Next, here, the divided signal correction circuits 130, 1
The histogram correction at 31 will be described with reference to FIG.

FIGS. 4A and 4B are explanatory diagrams of histogram correction in general video signal processing. Figure 4
(A) is the histogram data of the input signal. In this case, the input signal having the maximum value of dmax is divided into 6 gradations (first gradation to 6th gradation), and the input signal range of each gradation is dx.
1 to dx6, and the appearance frequency (histogram) of the signal of each gradation is H1 to H6. FIG. 4B is a correction for emphasizing the contrast of the grayscale signals (in this case, the 2nd grayscale and the 3rd grayscale) that appear frequently from the histogram data of FIG. 4A. In order to perform this correction, a different correction gain is applied to each gradation signal, for example, HG is applied to the second gradation signal.
The gain of 2 = out2 / in2 and the signal of the third gradation are multiplied by the gain of HG3 = out3 / in3, respectively.

Next, FIGS. 5A, 5B and 5C are shown in FIG.
6 is an explanatory diagram of histogram correction in the histogram correction circuits 305 and 306 shown in FIG. FIG. 5A shows AD-
Histogram data of the input signal of CH1, FIG. 5B is the histogram data of the input signal of AD-CH2,
Similarly to FIG. 4A, the input signal having the maximum value of dmax is divided into 6 gradations (first gradation to 6th gradation), the input signal range of each gradation is dx1 to dx6, and each floor The appearance frequency (histogram) of the key signal is H11 to H16 and H
21 to H26 (both solid lines in the figure).
Further, in FIG. 5B, the histogram data of AD-CH1 is shown by a dotted line in the figure. Here, the histogram data of AD-CH1 and AD-CH2 changes due to the variation in the difference between the channels, and the appearance frequency of each gradation is H11 <H.
21, H12> H22, H13> H23, H14 <H2
4, H15 = H25, H16 = H26. FIG. 5C shows each of the gradation data of AD-CH2 for matching the histogram data (H21 to H26) of each gradation of AD-CH2 with the histogram data (H11 to H16) of each gradation of AD-CH1. This is a correction gain characteristic. Here, the correction gain (HG2
1) is a positive gain, correction gains (HG22, HG23) for the second and third gradations are negative gains, correction gains for the fourth gradation (HG24) are positive, fifth and sixth gradations The correction gains (HG25, HG26) of 1 indicate the case of the gain of 1 time. Here, plus and minus mean whether it is larger or smaller than 1.

Next, in order to match this AD-CH2 histogram with the AD-CH1 histogram, AD-C
An example of the method of calculating the gain for each gradation for H2 is shown in FIG.
The case of (a) and (b) is shown. AD-CH1 and AD-C
The histogram data value of the first gradation of H2 is H
11, H21 (H11 <H21). This is essentially
It is considered that the signal that should be in the second gradation of AD-CH2 has changed to the signal level of the first gradation. This change is AD
When it is caused by a change in gain in the signal path of -CH1 and the signal path of AD-CH2, it means that the gain of the signal path of AD-CH2 is lower than the gain of the signal path of AD-CH1. Therefore, in order to make the histogram data value of the first gradation of AD-CH2 equal to H11, the signal within the first gradation of AD-CH2 is multiplied by a positive gain.

Next, an example of the method of calculating the gain value will be shown. If it is considered that the appearance frequencies of the respective signal levels are all equal in the range of signal levels within the same gradation, AD-C
Correction gain HG21 = for the signal within the first gradation of H2 =
If you multiply the gain by (H21 / H11), AD-CH2
The histogram data value of the first gradation of is equal to H11.

Next, a method of calculating the correction gain HG22 will be described. H for the signal in the first gradation of the AD-CH2
By the gain correction of G21 = (H21 / H11), A
The histogram data values of the second gradation of D-CH1 and AD-CH2 are H12, (H21-H11 + H22). Therefore, H12 is compared with (H21−H11 + H22) to perform gain correction so that the histogram data values of the second gradation of AD-CH1 and AD-CH2 are equal. For example, in the case of H12 <(H21−H11 + H22), the same processing as the first gradation is performed, and ((H21−H11 + H2) is applied to the signal in the second gradation of AD-CH2.
2) Multiply the positive gain of / H12). Also, H
In the case of 12> (H21-H11 + H22), contrary to the first gradation, the signal originally supposed to be at the second gradation of AD-CH2 has changed to the signal level at the third gradation. Conceivable. Therefore, in order to make the histogram data value of the second gradation of AD-CH2 equal to H12, AD-CH
The negative gain is applied to the signals in the second and third gradations of 2. This gain value can be calculated as follows.

First, the second of AD-CH1 and AD-CH2
If the difference between the histogram data values of the gradation is dh2,
dh2 = (H12- (H21-H11 + H22)), and the data for dh2 is gained down from the signal within the third gradation of AD-CH2 to the signal within the second gradation. Next, the AD-C corresponding to the data of dh2
If the signal range within the third gradation of H2 is dx23, dx
23 = dh2 / (H23 / dx3), and AD-CH
Gain (dx2 / (dx2 + dx23)) for converting a signal in the range (dx2 + dx23) of 2 into a signal in the range dx2
Is the correction gain HG22.

Thus, first, AD-CH1 and AD-
The correction gain HG21 for equalizing the histogram data value of each first gradation of CH2 is calculated, and then the histogram of each second gradation of AD-CH2 and AD-CH1 after the processing of the correction gain HG21. The correction gain HG22 for equalizing the data values is calculated, and the correction gain HG2 is further calculated.
1 and AD-CH2 and AD-CH1 after HG22 treatment
The correction gain HG23 for equalizing the histogram data values of the respective third gradations is calculated, and HG2
A correction gain for equalizing the respective histogram data values is calculated from the gradations of 4, HG25, HG26 and the low signal level.

In this way, the histogram detection circuit 20
5, 206 detects the histogram data of the signal of each channel, and the histogram correction circuits 305, 306 perform gain correction for each gradation so that the histogram data match.

The histogram detection circuits 205 and 206
In, the input signal is divided into a plurality of gradations, the appearance frequency (histogram) for each gradation is detected, and an error is detected from the detected histogram data of each channel. Therefore, for example, all pixels in the effective detection area are detected. It is possible to detect an error between channels with higher accuracy than detecting an error from the average value data. Further, by increasing the number of divisions of this gradation, it is possible to detect an error between channels with higher accuracy.

Further, the inter-channel detection circuit 113 shown in FIG. 2 and the divided signal correction circuits 130 and 13 shown in FIG.
The configuration example of No. 1 shows the configuration in which all the processes are performed in the inter-channel detection and the division signal correction, but it is not necessarily the configuration in which all the processes are performed. There may be a configuration in which two processes are performed.

Next, a specific example of correction between channels will be described. FIG. 6 shows a setting example of the detection area setting circuit 201.
FIG. 6A shows a case where the vicinity of the division boundary of the image data of the image pickup area is set to the left and right specific positions (dotted line areas in the figure) in the horizontal direction. In this case, the color of the image pickup element is set in the horizontal direction. Set an integer multiple of the filter repetition period. In FIG. 6B, an OB area (hatched area in the drawing) that is an optical black level area is provided in the image pickup area of the image pickup element, and the OB area and the image data are identified in the lateral direction from the vicinity of the division boundary. This is a case where up to the position (dotted line area in the figure) is set.

An example of processing for detecting and correcting a difference between channels in the case of each setting example of the detection area setting circuit 201 shown in FIGS. 6A and 6B is shown in FIG.
6 (b) will be described with reference to the flowcharts shown in FIGS.

First, in the case of FIG. 6A, since the difference between channels is detected using the image data of the image pickup area, it is necessary to avoid the influence of the subject in the detection area near the division boundary. Therefore, the reliability area detection circuit 202 evaluates the reliability of AD-CH1 and AD-CH2, detects a reliable area, and sets information on the reliable area in each detection circuit. Here, as a signal that is not reliable and is not suitable for use in detecting the difference between channels,
For example, there are 1) a signal of a saturated region, 2) a signal of a boundary region of different objects, and the like. As a method of detecting a signal of these regions, 1) each of AD-CH1 and AD-CH2
Whether the signal level of is within a certain range, 2) A
It depends on whether the signal level difference between D-CH1 and AD-CH2 exists within a certain range. In this way, the signal in the unreliable area is removed, and the difference between the channels is detected using the reliable area indicating signal.

Next, the average value detection circuits 203 and 204 and the histogram detection circuits 205 and 20 in each channel.
The operation of No. 6 will be described. Average value detection circuits 203, 2
In FIG. 4, the detection area setting circuit 201 sets FIG.
Using the reliable area indicating signal detected by the reliability area detecting circuit 202 in the detection area (dotted line area in the figure) shown in (a), the average value of the signals corresponding to the black level and the average value of the entire signals are calculated. To detect. In the histogram detection circuits 205 and 206, the reliability area detection circuit 2 is set in the detection area (dotted area in the drawing) shown in FIG. 6A set by the detection area setting circuit 201.
The signal level distribution is detected using the reliable area indication signal detected by 02. Peak detection circuits 230, 23
6A set by the detection area setting circuit 201 in FIG.
Using the reliable area indication signal detected by the reliability area detection circuit 202 within the detection area (dotted line area in the figure) indicated by
Detect signal level distribution. Using the detected data, the overall control circuit 114 shown in FIG. 1 obtains the correction data of the difference between the channels, the offset correction circuits 301 and 302 shown in FIG. 3 correct the offset error, and the gain correction circuit 303, In 304, gain correction is performed using the peak level data, and in histogram correction circuits 305 and 306, gain correction is performed using the histogram data.

FIG. 7 shows an example of the processing flow in the case of FIG. 6 (a). As shown in FIG. 7, the detection area setting circuit 201 first sets the detection area (dotted line area in the drawing) shown in FIG. 6A as the detection area (process 601), and then the reliability area detection circuit 202. Detects a reliable area in this detection area (process 602). In this reliable area, the average value detection circuits 203 and 204 detect the area corresponding to the black level from the signals of AD-CH1 and AD-CH2 (process 603), and the average of the signals of the detection area corresponding to the black level is calculated. By detecting the value (process 604), the offset correction value for correcting the black level is calculated in the overall control circuit 114 (process 605).

In the area with reliability, AD
The histogram detection circuits 205 and 206 detect the histogram of the division boundary area from the signals of -CH1 and AD-CH2 (process 606), and use this histogram data and the offset correction data obtained in the process 605 to perform overall control. In the circuit 114, a histogram correction value for correcting the histogram is calculated (process 607).

Although the case where the offset correction and the histogram correction are performed has been described with reference to FIG. 7, there are cases where the offset correction and the gain correction are performed.

Next, the case of FIG. 6B will be described focusing on the points different from FIG. 6A. In the case of FIG. 6B, the inter-channel difference is detected using the image data of the imaging area and the signal of the OB area which is an optical black level area. In the detection using the image data, it is necessary to avoid the influence of the subject in the detection area near the division boundary as in the case of FIG. 6A, and AD-CH1,
The reliability of the signal of AD-CH2 is evaluated, the reliable area is detected, the information of the reliable area is set in each detection circuit, and each detection circuit uses the signal of this area to perform channel detection. The difference is detected.

The average value detection circuits 203 and 204 use the signals of the OB upper area and the OB lower area (hatched area in the figure) set by the detection area setting circuit 201 as shown in FIG. Of the entire signal is detected by using the reliable area indicating signal detected by the reliability area detecting circuit 202 in the image detection area (dotted area in the figure) which is the image data. To detect. In the histogram detection circuits 205 and 206, the image detection area set by the detection area setting circuit 201 shown in FIG.
The distribution of the signal level is detected using the reliable area indicating signal detected by the reliable area detecting circuit 202. In the peak detection circuits 230 and 231, the detection area setting circuit 2
The peak level of the signal is detected using the reliable area indication signal detected by the reliability area detection circuit 202 within the image detection area (dotted area in the drawing) shown in FIG. 6B set by 01. Using the detected data, the overall control circuit 114 shown in FIG. 1 obtains the correction data of the difference between the channels, and the offset correction circuits 301 and 302 shown in FIG.
The offset error is corrected in, the gain error is corrected in the gain correction circuits 303 and 304, and the gain error is corrected using the histogram data in the histogram correction circuits 305 and 306.

The processing flow in the case of FIG. 6B is shown in FIGS.

As shown in FIG. 8, first, the detection area setting circuit 201 sets the detection area (hatched area and dotted area in the figure) shown in FIG. 6B as the detection area (process 601),
Next, the average value of the signal in the OB area (hatched area in the figure) corresponding to the black level is detected (process 608) to calculate the offset correction value for correcting the black level in the overall control circuit 114 (process 605). To do.

Further, the reliability area detection circuit 202 detects a reliable area in this detection area (dotted line area in the figure) (process 602). In this reliable region, the histogram detection circuits 205 and 206 detect the histogram of the division boundary region from the signals of AD-CH1 and AD-CH2 (process 606), and this histogram data and the offset correction obtained in the process 605 are performed. Using the data, the overall control circuit 114 calculates a histogram correction value for correcting the histogram (process 607).

As described above, in FIG. 6B, by having the OB area which is an optical black level area, the black level is detected and the black level is corrected as compared with the case where the data of the image detection area is used. The offset correction can be performed with high accuracy.

Although the case where the offset correction by the data of the OB area and the histogram correction by the data of the image detection area are performed has been described with reference to FIG. 8, the gain correction may be performed from the peak data of the image detection area.

In the inter-channel detection circuit 113 shown in FIG. 2 in FIGS. 7 and 8, the reliability area detection circuit 202 detects a reliable area in the detection area (dotted line area in the drawings), The case where the average value detection circuits 203 and 204, the histogram detection circuits 205 and 206, and the peak detection circuits 230 and 231 obtain the average value data, the histogram data, and the peak data for each channel from the detected reliable area has been described. However, it is also possible to judge the reliability of these detection data by using average value data, histogram data, and peak data in each channel detected by these detection circuits. A processing example in this case will be described with reference to FIG.

As shown in FIG. 9, first, the detection area setting circuit 201 sets the detection area (hatched area and dotted area in the figure) shown in FIG. 6B as a detection area (process 601),
Next, the average value of the signal in the OB area (hatched area in the figure) corresponding to the black level is detected (process 608) to calculate the offset correction value for correcting the black level in the overall control circuit 114 (process 605). To do.

Further, the reliable area detection circuit 202 detects a reliable area by detecting a saturated area in the OB area (hatched area in the figure), and then detects the detected reliable area. Histogram data of each channel in the area (1) is detected (process 630), and the signal difference between the channels in the reliable area may be due to the difference in the processing of the path for each channel or the difference between the subjects. It is determined (process 631).

If the result of this judgment is due to the difference in the processing of the paths for each channel, it is judged that the detection data from the image detection area (dotted line area in the figure) is reliable, and the average value data or peak The data is detected (process 632), and the average value data or peak data and the process 6 are processed.
Using the offset correction data obtained in step 05, the overall control circuit 114 calculates a gain correction value (process 63).
4) Do. In addition, if the result of the determination is due to the difference between the subjects, it is highly likely that an erroneous correction will occur if the correction between channels is performed using the data detected from the signal of the image detection area (dotted line area in the figure). Image detection area (dotted line area in the figure)
It is judged that the reliability of the detection data from is unreliable, and this unreliable detection data is not used, and as reliable detection data, for example, the image detection area immediately before the reliability is judged (dotted line area in the figure) Using the data detected from the signal (process 633) and using the immediately preceding reliable average value data or peak data and the offset correction data obtained in the process 605, the overall control circuit 114 gain correction A value is calculated (process 634). Alternatively, only the offset correction based on the offset correction value detected from the OB area, which is reliable detection data, is performed.

As described above, in FIG. 6B, by having the OB area which is an optical black level area, the black level is detected and the black level is corrected as compared with the case where the data of the image detection area is used. The offset correction can be performed with high accuracy, and the reliability of the data detected from the image detection area is determined by using the predetermined data detected from the image detection area of each channel. It is possible to perform highly accurate correction.

In FIG. 9, the reliability of the data detected from the image detection area is judged by using the histogram data from the image detection area, and the gain correction is performed by the reliable average value data or peak data. As described above, it is also possible to select average value data and peak data by using histogram data. As the selection method, for example, in the detection of histogram data, the gradation width near the high level of the input signal is narrowed to improve the detection accuracy of peak data, and the peak level value and its appearance frequency are compared between channels to determine the peak. It is conceivable to judge the reliability of the data and use the peak data when the reliability of the peak data is high, and use the average value data when the reliability of the peak data is low.

Further, as in the case where the reliability of the data detected from the image detection area is judged by using the peak data from the image detection area and the gain correction is performed by the reliable average value data or histogram data, The detection data used for the reliability determination and the detection data used for the correction data may be other combinations including the case where the same detection data is used for the reliability determination and the correction data.

Further, although the case where the offset correction value detected from the OB area is used to calculate the gain correction value has been described with reference to FIG. 9, the image detection area is used when the reliability is judged by the data detected from the image detection area. OB for the signal of
Performs correction using the offset correction value detected from the area,
It is also possible to further improve the accuracy of the reliability judgment by judging the reliability from the corrected signal.

As described above, in the first embodiment,
Histogram detection circuits 205 and 206 are provided. By detecting the histogram data of the signals of the respective regions divided by the histogram detection circuits 205 and 206, the error between channels is detected with high accuracy. , 305, it is possible to correct the error between the channels in the histogram detection circuits 304, 305 by calculating a histogram correction value for matching the histogram data for each channel.

Further, the reliability of the data from the image detection area is judged by using the data from the image detection area, and the reliability of the error detection between the channels is improved by using the reliable data. It is possible.

(Second Embodiment) FIG. 10 is a block diagram showing the arrangement of an image pickup apparatus according to the second embodiment of the present invention. Hereinafter, the points different from the first embodiment will be mainly described.

Referring to FIG. 10, the first embodiment shown in FIG.
The same reference numerals are given to the portions corresponding to those of the configuration. The divided signal correction circuit is composed of offset correction circuits 107 and 108 and histogram correction circuits 109 and 110, and the input signal to the inter-channel detection circuit 113 is offset correction circuits 107 and 108 and histogram correction circuit 10.
It is different from the first embodiment in that it is a signal after 9,110. The other configurations are similar to those of the first embodiment, and therefore the description thereof is omitted.

The operation of the image pickup apparatus configured as described above will be described below.

In FIG. 10, the A / D conversion circuit 105 and the A / D conversion circuit 1 obtained by the same processing as in FIG.
The output signal of 06 is an offset correction circuit 1 which is a correction means for correcting the difference between the channels of both output signals.
07 and 108, and these offset correction circuits 1
At 07 and 108, offset correction processing of both signals is performed using the other input signal, which is the predetermined offset correction control signals OF1 and OF2 calculated and set by the overall control circuit 114. Next, the output signals of the offset correction circuits 107 and 108 are input to the histogram correction circuits 109 and 110, which are another correction means for correcting the difference between the channels of both output signals, and these histogram correction circuits 109 and 110 are input. At 110, the histogram correction processing of both signals is performed using the other input signal, which is the predetermined histogram correction control signals HG1 and HG2 calculated and set by the overall control circuit 114.

Further, the histogram correction circuits 109 and 11
The output signals of AD-CH1 and AD-CH2, which are output signals of 0, are simultaneously input to the inter-channel detection circuit 113, which is detection means for detecting the difference between the channels of both output signals, and the inter-channel detection circuit 113 is detected. The detection data detected in (4) is input to the overall control circuit 114.

As described above, the output signals from the offset correction circuit and the histogram correction circuit, which are the correction means for correcting the difference between the channels, are the inter-channel detection circuit 113.
Then, the difference between the channels is detected, and the correction control signal calculated by the overall control circuit 114 from the data of the difference between the channels is fed back to the correction means. Here, the correction of the difference between the channels may be performed only by the offset correction or the histogram correction depending on the case.

In this way, the image data output in which the difference between the channels generated between the two outputs is corrected is converted into one image data by the image synthesizing circuit 111 (the left and right outputs are made into one output), Next camera signal processing circuit 1
At 12, predetermined signal processing (color interpolation processing, γ conversion, etc.) is performed.

Next, with respect to the detection and correction of the difference between channels by this feedback configuration, the case of histogram correction will be described. The histogram correction circuits 109 and 110 shown in FIG. 10 are similar to the histogram correction circuits 305 and 306 shown in FIG. 3, and the inter-channel detection circuit 113 shown in FIG. Similar to the detection circuit 113, it is shown in FIG. Less than,
Operations of the histogram detection circuits 205 and 206 shown in FIG. 2 and the histogram correction circuits 109 and 110 shown in FIG. 10 will be described with reference to FIG.

FIG. 5A shows the histogram data of the input signal of AD-CH1, and FIG. 5B shows the histogram data of the input signal of AD-CH2. The maximum value of the input signal is dmax for 6 gradations ( Divided from the first gradation to the sixth gradation),
The input signal range of each gradation is dx1 to dx6, and the appearance frequency (histogram) of the signal of each gradation is H11 to
H16 and H21 to H26 (both solid lines in the figure) are shown. Further, in FIG. 5B, the histogram data of AD-CH1 is shown by a dotted line in the figure. Here, due to variations in channel differences, AD-CH1 and AD-CH
The histogram data of No. 2 changes, and the appearance frequency for each gradation is H11 <H21, H12> H22, H13> H23,
The case where H14 <H24, H15 = H25, H16 = H26 is shown. FIG. 5C shows AD histogram data (H21 to H26) for each gradation of AD-CH2.
It is a correction gain characteristic for each gradation of AD-CH2 for matching with the histogram data (H11 to H16) for each gradation of -CH1, and here the correction gain (HG21) of the first gradation is a positive gain. , The second and third gradation correction gains (HG22, HG23) are negative gains, the fourth gradation correction gain (HG24) is positive, the fifth and sixth gradation correction gains (HG25, HG26) shows the case of a gain of 1 time.

Next, in order to match this AD-CH2 histogram with the AD-CH1 histogram, AD-C
An example of the method of calculating the gain for each gradation for H2 is shown in FIG.
The case of (a) and (b) is shown. AD-CH1 and AD-C
The histogram data values of the first gradation of H2 are H11, H
21 (H11 <H21). This is originally AD-C
It is considered that the signal that should be in the second gradation of H2 has changed to the signal level of the first gradation. This change is AD-CH1
In the case of the change in the gain of the signal path of AD-CH2 and the gain of the signal path of AD-CH2, it means that the gain of the signal path of AD-CH2 is lower than the gain of the signal path of AD-CH1. Therefore, in order to make the histogram data value of the first gradation of AD-CH2 equal to H11, A
A signal within the first gradation of D-CH2 will be multiplied by a positive gain. Next, an example of a method of calculating the gain value will be shown. When it is considered that the appearance frequencies of the respective signal levels are all equal in the range of the signal levels within the same gradation, A
Correction gain HG2 for the signal within the first gradation of D-CH2
1 = (H21 / H11) gain, AD-C
The histogram data value of the first gradation of H2 becomes equal to H11.

As described above, first, the correction gain (HG2
1) is calculated, and the signal in at least the first gradation and the second gradation of AD-CH2 is multiplied by the correction gain. Next, the histogram data of AD-CH2 after this correction gain (HG21) is detected by the histogram detection circuit 206 and compared with the histogram data of AD-CH1. In the comparison of the histogram data, the histogram data values of the first gradation are substantially coincident with each other within a predetermined range by the correction gain control for the signal of the first gradation of AD-CH2. -CH1 and AD-CH2 second
The gain correction is performed so that the histogram data values of the gradations become equal.

The histogram data value of the second gradation of AD-CH2 after the gain correction of HG21 for the signal in the first gradation of AD-CH2 is set to H22 ', for example, H
When 12 <H22 ′, the same processing as the first gradation is performed, and the signal within the second gradation of AD-CH2 is multiplied by the positive gain of (H22 ′ / H12). Further, in the case of H12> H22 ′, it is considered that, contrary to the first gradation, the signal originally supposed to be at the second gradation of AD-CH2 is changed to the signal level at the third gradation. To be Therefore, in order to make the histogram data value of the second gradation of AD-CH2 equal to H12, it is necessary to multiply the signals in at least the second gradation and the third gradation of AD-CH2 by a negative gain. Become. This gain value can be calculated as follows. First, if the difference between the histogram data values of the second gradation of AD-CH1 and AD-CH2 is dh2, then dh2 = (H12-H22 '), and this dh2
Minute data from the signal within the third gradation of AD-CH2 to the second
The gain is reduced to the signal within the gradation. Next, when the signal range within the third gradation of AD-CH2 corresponding to the data for dh2 is dx23, dx23 = dh2 /
(H23 / dx3), and (dx2 +) of AD-CH2
The gain (dx2 / (dx2 + dx23)) for converting the signal in the range of dx23) into the signal in the range of dx2 is the correction gain H.
It becomes G22.

Thus, first, AD-CH1 and AD-
The correction gain HG21 for equalizing the histogram data values of the respective first gradations of CH2 is calculated, and then the histogram data of AD-CH2 after the processing of the correction gain HG21 is detected, and AD-CH2 and AD- The correction gain HG22 for equalizing the histogram data values of the respective second gradations of CH1 is calculated, and the histogram data of AD-CH2 after the processing of the correction gain HG22 is detected, and AD
The correction gain HG23 for equalizing the histogram data values of the third gradations of -CH2 and AD-CH1 is calculated, and similarly, the histogram data values are calculated from HG24, HG25, HG26 and the gradation of the low signal level. Then, a correction gain for equalizing is calculated.

As described above, the histogram data of the signal of each channel is detected by the histogram detection circuits 205 and 206, and the gain correction is performed for each gradation by the histogram correction circuits 109 and 110 so that the histogram data match. After that, the histogram data of the signal after the gain correction is detected by the histogram detection circuits 205 and 206, and the gain correction is performed for each gradation in the histogram correction circuits 109 and 110 so that the histogram data match.

In the above, the first of AD-CH2 is
The case where the histogram data value of the first gradation substantially matches the histogram data value of the first gradation of AD-CH1 within a predetermined range by the correction gain (HG21) control for the gradation signal will be described. However, if the histogram data of the first gradation is compared again and the value is out of the predetermined range, the correction gain HG21 ′ for equalizing the detected histogram data value of the first gradation is calculated. do it,
Correction gain (H for the first gradation signal of AD-CH2
G21 ′) control allows the histogram data value of the first gradation to be repeatedly corrected until it substantially matches the histogram data value of the first gradation of AD-CH1 within a predetermined range.

As described above, in the second embodiment,
Histogram detection circuits 205 and 206 are provided. By detecting the histogram data of the signals of the respective regions divided by the histogram detection circuits 205 and 206, the error between channels is detected with high accuracy, and the histogram correction circuit 109 is further provided. , 110, and a histogram correction value for matching the histogram data for each channel is calculated, and the error between the channels is repeatedly corrected in the histogram correction circuits 109 and 110, so that it is possible to perform the accurate inter-channel correction. Is.

(Third Embodiment) FIG. 11 is a block diagram showing the structure of an inter-channel detection circuit of an image pickup apparatus according to the third embodiment of the present invention. Hereinafter, the points different from the first and second embodiments will be mainly described.

Referring to FIG. 11, the first embodiment shown in FIG.
The same reference numerals are given to the portions corresponding to those of the configuration. The difference is that the peak detection circuits 230 and 231 are not provided, but all-channel average value detection circuit 207 and all-channel histogram detection circuit 208 are added instead.

With respect to the image pickup apparatus constructed as described above, the operation of detecting between channels will be described below.

In FIG. 11, reference numeral 201 designates FIGS.
A detection area setting circuit that sets a detection area under the control of the overall control circuit 114 shown in FIG. 2, 202 is a reliability area detection circuit that detects an area with reliability for an input signal, 20
3, 204 is an average value detection circuit for detecting the average value of the input signal in each channel in the set range, 205,
206 is a histogram detection circuit that detects the histogram (signal level distribution) of the input signal in each channel in the set range, 207 is the average value of all channels that detects the average value of the input signals in both channels in the set range A detection circuit 208 is an all-channel histogram detection circuit that detects the histogram (signal level distribution) of the input signals of both channels in the set range. Hereinafter, the operation of the inter-channel detection circuit thus configured will be described.

In FIG. 11, first, the detection area setting circuit 201 sets an area for detecting a difference between channels. This detection area setting circuit 201 is shown in FIG.
0 from the timing signal generation circuit 102 shown in FIG.
With the HD signal as a reference, the effective range for the output signal from the image sensor 101 is determined, and the timing of detection by each detection circuit is set. Further, the reliability area detection circuit 202 evaluates the reliability of AD-CH1 and AD-CH2, detects a reliable area, and sets information on the reliable area in each detection circuit. Next, the average value detection circuits 203 and 204 in each channel use the reliable area indication signal detected by the reliability area detection circuit 202 in the detection area set by the detection area setting circuit 201, and All channel average value detection circuit 2
In 07, using the signals of the detection areas of the reliable average value detection circuits 203 and 204 detected by the reliability area detection circuit 202 in the detection area set by the detection area setting circuit 201, for example, the black level equivalent The average value of the signal and the average value of the entire signal are detected. In addition, the histogram detection circuit 2
In 05 and 206, the reliable area indication signal detected by the reliability area detection circuit 202 in the detection area set by the detection area setting circuit 201 is used, and in the all-ch histogram detection circuit 208, the detection area setting circuit The reliable histogram detection circuits 205 and 206 detected by the reliability area detection circuit 202 within the detection area set by 201.
The signal level distribution is detected using the signal in the detection area of. By using these detected data, the correction data of the difference between the channels is obtained in the overall control circuit 114 shown in FIGS.

Next, an example of a method of calculating the correction data of the difference between channels will be described below with reference to FIG.

FIG. 6 shows a setting example of the detection area setting circuit 201, and FIG. 6A shows setting from the vicinity of the division boundary of the image data of the image pickup area to the left and right specific positions (dotted line areas in the figure) in the horizontal direction. This is the case, and in this case, an integer range of the repetition cycle of the color filter of the image sensor is set in the horizontal direction.

First, in the case of FIG. 6A, since the difference between channels is detected by using the image data of the image pickup area, it is necessary to avoid the influence of the subject in the detection area near the division boundary. Therefore, the reliability area detection circuit 202 evaluates the reliability of AD-CH1 and AD-CH2, detects a reliable area, and sets the information of the reliable area in each detection circuit. The difference between channels is detected using the presence area indicating signal.

Next, the operations of the average value detection circuit and the histogram detection circuit will be described. Average value detection circuit 203,
At 204, the detection area setting circuit 201 sets the
Using the reliable area indicating signal detected by the reliability area detecting circuit 202 in the detection area (dotted line area in the figure) shown in (a), the average value of the signals corresponding to the black level and the average value of the entire signals are calculated. To detect. Further, in the all-channel average value detection circuit 207, the reliability average value detection circuit 203 which is detected by the reliability area detection circuit 202 in the detection area set by the detection area setting circuit 201 and shown in FIG. Similarly, the average value of the signals corresponding to the black level and the average value of the entire signals are detected by using both the signals of the detection areas 204.

On the other hand, the histogram detection circuits 205, 20
6, the detection area setting circuit 201 sets FIG.
Using the reliable area indication signal detected by the reliability area detection circuit 202 within the detection area (dotted line area in the figure) indicated by
Detect signal level distribution. Further, in the all-channel histogram detection circuit 208, the reliable histogram detection circuits 205 and 206 detected by the reliability area detection circuit 202 within the detection area set by the detection area setting circuit 201.
The signal level distribution is detected using both of the signals in the detection area.

Using the detected data, correction data for the difference between channels is obtained in the overall control circuit 114 shown in FIGS. 1 and 10, and the offset correction circuits 107 and 10 are used.
8, the offset error is corrected, and the histogram correction circuits 109 and 110 perform the gain correction using the histogram data.

The processing flow in the case of FIG. 6A is shown in FIG.
2 shows. In FIG. 12, parts corresponding to those in the process shown in FIG. 7 are designated by the same reference numerals. As shown in FIG. 12, first, the detection area setting circuit 201 sets the detection area (dotted line area in the figure) shown in FIG. 6A as a detection area (process 601), and then the reliability area detection circuit 202. Detects a reliable area within this detection area (process 60).
2) Do.

In this reliable area, AD-CH
1 and AD-CH2 signals, average value detection circuit 203
And 204 detect a region corresponding to the black level (process 603)
Then, the average value of the signals in the detection area corresponding to the black level is detected (process 604). In addition, the average value detection circuit 20 for all channels
7 detects the areas corresponding to the black levels of the signals of both channels (processing 903), and detects the average value of the signals of the detection areas corresponding to the black levels (processing 904), so that the overall control circuit 114 receives each channel. An offset correction value (OF1 and OF2 in FIGS. 1 and 10) for correcting the black level of 1 to the black level detected from both channels is calculated (process 905).

In addition, in the above-mentioned area with reliability, AD
From the -CH1 and AD-CH2 signals, the histogram detection circuits 205 and 206 detect the histograms of the division boundary regions (process 606), and the all-ch histogram detection circuit 208 detects the histograms of the division boundary regions of both channels. (Process 906) Then, by using these histogram data and the offset correction data obtained in the process 905, a histogram for correcting the histogram of each channel in the overall control circuit 114 into the histogram of the division boundary region detected from both channels. Correction values (HG1 and HG2 in FIGS. 1 and 10) are calculated (process 907).

As described above, the black level average value and the histogram detection value obtained from the regions of the respective channels are corrected so as to match the black level average value and the histogram detection value obtained from the regions of both channels, thereby obtaining a match. Since the area for calculating the data to be expanded becomes wide, highly accurate target data can be obtained, and since the signals of both channels are individually corrected, the content of the signal processing of each channel is common, It is possible to perform highly accurate correction without causing a difference in signal of each channel due to a difference in signal processing.

As described above, in the third embodiment,
An all-channel average value detection circuit 207 and an all-channel histogram detection circuit 208 are provided, and by detecting data from the signals of the regions of both channels in these detection circuits, the error between the channels is detected with high accuracy. It is possible to accurately correct an error between channels.

(Fourth Embodiment) FIG. 13 is a block diagram showing an example of the hardware configuration of an image pickup apparatus according to the fourth embodiment of the present invention. Hereinafter, the points different from the first embodiment will be mainly described.

Referring to FIG. 13, the first embodiment shown in FIG.
The same reference numerals are given to the portions corresponding to those of the configuration. A big difference is that a pilot charge corresponding image pickup device 116 having a region for storing a predetermined charge is provided and a case where the divided signal correction circuit is composed of the offset correction circuits 107 and 108 and the gain correction circuits 118 and 119. That is the point.

In FIG. 13, reference numeral 116 designates a plurality of (here, two) image pickup regions, a region for storing a predetermined charge (hereinafter, also referred to as pilot charge), and a signal for these plurality of image pickup regions and a region for storing the predetermined charge. An image pickup device such as a CCD, which is an image pickup unit for independently outputting (CH1 and CH2), 117 is an image pickup device 1 for pilot charge
A timing signal generating circuit for generating timing signals necessary for driving and signal processing of 16; 103, 104 are analog signal processing circuits for performing noise removal, gain control, etc., 10
Reference numerals 5 and 106 denote analog-to-digital conversion circuits (hereinafter referred to as A / D conversion circuits) that convert analog signals to digital signals, and 107 and 108 denote offset components of an input signal which are correction means for a difference between channels of two signals. Offset correction circuits for correction, 118 and 119 are gain correction circuits that correct the gain of an input signal, which is a means for correcting the difference between channels of two signals, 111 is an image synthesis circuit that synthesizes a plurality of video signals, and 112 is a camera signal. Processing circuit, 120
Is an inter-channel detection circuit for detecting a difference between a plurality of signals, 12
Reference numeral 1 is an overall control circuit for controlling the above circuit, and 122 is a mode setting circuit.

The operation of the image pickup apparatus configured as described above will be described below.

In FIG. 13, the pilot charge-corresponding image pickup device 116 divides the subject image into a plurality of image pickup regions by the drive signal from the timing signal generation circuit 117, picks up an image, and outputs two signals from each of the plurality of image pickup regions ( C
H1 and CH2), and the signals in the pilot charge region are two signals (CH) corresponding to a plurality of imaging regions.
1 and CH2). The data output from the right half surface of the image sensor 116 corresponding to pilot charge is CH1.
It is input to the analog signal processing circuit 104 via the output, and by performing a known method such as correlated double sampling here, reset noise and the like included in the output of the CCD and the like are removed and the signal is output up to a predetermined signal level. Activates the AGC circuit for amplifying. The output after this AGC is A / D
By inputting to the conversion circuit 106, it is converted into a digital signal and an output of AD-CH1 is obtained.

Similarly, the data output from the left half surface of the image pickup device is sent to the analog signal processing circuit 10 via the CH2 output.
3 and the same method such as correlated double sampling is performed here to remove the reset noise and the like contained in the output of the CCD and the like, and an AGC circuit for amplifying the output to a predetermined signal level. To work. By inputting the output after this AGC to the A / D conversion circuit 105, it is converted into a digital signal and an output of AD-CH2 is obtained.

Next, the output signals of AD-CH1 and AD-CH2 are simultaneously input to the inter-channel detection circuit 120 which is a detection means for detecting the difference between the channels of both output signals, and the inter-channel detection circuit 120 is detected. The detection data detected by is input to the overall control circuit 121.

The output signals of AD-CH1 and AD-CH2 are offset correction circuits 107 and 108 which are correction means for correcting the difference between the channels of both output signals.
Is input to these offset correction circuits 107 and 108.
Then, the predetermined offset correction control signal OF1, which is calculated and set by the overall control circuit 121, which is the other input signal.
Offset correction processing of both signals is performed using OF2. Next, the output signals of the offset correction circuits 107 and 108 are
The signals are input to gain correction circuits 118 and 119 that are another correction means for correcting the difference between the channels of both output signals. In these gain correction circuits 118 and 119, the overall control circuit 121 that is the other input signal is input. The gain correction processing of both signals is performed using the predetermined gain correction control signals GA1 and GA2 calculated and set in step 1.

As described above, the two output signals are detected by the inter-channel detection circuit 120 and the overall control circuit 12 is detected.
The difference between the channels is corrected by the offset correction circuits 107 and 108 and the gain correction circuits 118 and 119 which are correction means for performing correction based on the control signal calculated in 1. Here, the correction of the difference between the channels may be performed only by the offset correction or the gain correction depending on the case.

In this way, the image data output in which the difference between the channels generated between the two outputs is corrected is converted into one image data by the image synthesizing circuit 111 (the left and right outputs are made into one output), and Stage camera signal processing circuit 11
In step 2, predetermined signal processing (color interpolation processing, γ conversion, etc.) is performed.

Next, the specific structure of the inter-channel detection circuit 120 will be described with reference to the structural example shown in FIG.

In FIG. 14, 209 is a detection area setting circuit for setting a detection area under the control of the overall control circuit 121 shown in FIG. 13, and 202 is a reliability area for detecting a reliable area for an input signal. Detection circuit, 203, 20
Reference numeral 4 is an average value detection circuit for detecting the average value of the input signal in the set range, 210 and 211 are pilot charge detection circuits for the input signal from the pilot charge region, and the inter-channel detection circuit thus configured will be described below. 1
The operation of 20 will be described.

In FIG. 14, first, the detection area setting circuit 209 sets an area for detecting a difference between channels. This detection area setting circuit 209 determines the effective range for the output signal from the pilot charge corresponding image pickup device 116 with reference to the VD / HD signal from the timing signal generation circuit 117 shown in FIG. Set the timing to do. Further, the reliability area detection circuit 202 evaluates the reliability of AD-CH1 and AD-CH2, detects a reliable area, and sets information on the reliable area in each detection circuit. Next, the average value detection circuits 203 and 204 in each channel use, for example, the black level equivalent by using the reliable area indication signal detected by the reliability area detection circuit 202 in the detection area set by the detection area setting circuit 209. The average value of the signal of and the average value of the whole signal are detected. Further, the pilot charge detection circuits 210 and 211 detect the signal due to the pilot charge by using the reliable area indication signal detected by the reliability area detection circuit 202 in the detection area set by the detection area setting circuit 209. Using these detected data, FIG.
The overall control circuit 121 shown in (1) obtains the correction data of the difference between the channels.

Next, a specific example of correction between channels will be described. 15 and 16 show setting examples of the detection area setting circuit 209. FIG. 15A shows an OB area (shaded area in the figure) which is an optical black level area on one side of the imaging area of the image sensor, and a pilot charge area (filled in the figure) in an area different from the image area of the image sensor. Region), and in this case, the pilot charge region is located outside the imaging region so that the pilot charge passes through all the paths of the left and right divided screens.

FIG. 15B shows an OB area (shaded area in the drawing) which is an optical black level area in the imaging area of the image sensor and a pilot charge area (filled in the figure) in an area different from the image area of the image sensor. Region), in which case the charges in the OB region and pilot charge region are output from the left and right divided regions. FIG. 16A shows the pilot charge region (filled region in the figure) located in a region different from the image pickup region of the image pickup device, as the outside (first pilot charge region) of the image pickup region shown in FIG. HCCD left screen and HC in HCCD that transfers charges in the horizontal direction
Between CD right screen and output amplifier (2nd pilot charge area)
In this case, the second pilot charge is the output amplifier (left screen) and the output amplifier which is each output amplifier of the left and right screens.
(Right screen) Differences in the subsequent paths, the first pilot charges detect differences in all paths of the left and right divided screens.

FIG. 16B shows an OB area (shaded area in the drawing) which is an optical black level area in the image pickup area of the image pickup device and the first pilot charge outside the image pickup area shown in FIG. 16A. There is a second pilot charge region between the area and the HCCD left screen and the HCCD right screen and the output amplifier. In this case, the second pilot charge is also the output amplifier (left screen) and the output amplifier which are the output amplifiers of the left and right screens. The difference (right screen) and the first pilot charge detect the difference in all paths of the left and right divided screens.

The detection and correction of the difference between channels in the case of each setting example of the detection area setting circuit 209 shown in FIGS. 15 and 16 will be described with reference to FIGS.
It demonstrates using the flowchart of the process shown to FIGS. 17-20 corresponding to (a) and (b).

First, in the case of FIG. 15A, the black level is detected using the signal of the OB area on the left screen side, and the signal of the pilot charge area outside the imaging area different from the imaging area of the imaging device is detected. To detect the difference between channels. Next, the detection and correction operation by the pilot charges in each channel will be described.

The average value detection circuit 203 detects the average value of the signals corresponding to the black level by using the signals in the OB area (hatched area in the figure) set by the detection area setting circuit 209 and shown in FIG. In the pilot charge detection circuits 210 and 211, a signal of the pilot charge area (filled area in the figure) shown in FIG. 15A set by the detection area setting circuit 209 is used to correspond to a predetermined amount of pilot charge. Detect the signal. At this time, at least two types of charge amounts are stored as pilot charges, and the data detected from the signals corresponding to the at least two types of charge amounts are used to correct the difference between the channels in the overall control circuit 121 shown in FIG. Offset correction value (OF1, OF
2) and gain correction values (GA1, GA2) are calculated, offset errors are corrected in offset correction circuits 107, 108, and gain correction of signal level is performed in gain correction circuits 118, 119.

FIG. 17 shows the processing flow in the case of FIG. 15 (a). As shown in FIG. 17, first, the detection area setting circuit 209 sets the OB area and the pilot charge area (hatched area and filled area in the figure) shown in FIG. 15A as the detection area (process 601). Next, the average value detection circuit 203 detects the average value of the signal corresponding to the black level from the signal in the OB area (in this case, output from AD-CH1) (step 608), and AD-in the pilot charge area
From the signals of CH1 and AD-CH2, the pilot charge detection circuits 210 and 211 detect the region of the first charge level (process 609) and the region of the second charge level (process 611) to detect the first charge level. The detection of the average value of the signal in the detection region (process 610) and the detection of the average value of the signal in the detection region of the second charge level (process 612) are performed. Using this detection data, the overall control circuit 121 calculates an offset correction value corresponding to the black level correction and a gain correction value corresponding to the signal level correction (process 613).

Next, an example of the method of calculating the correction value will be described. AD-CH1 and AD-CH2 for a predetermined charge amount c in the pilot charge corresponding image pickup device 116
Let V1 = G1 * f (c) + O1 V2 = G2 * f (c) + O2, respectively, and let the output signal of the OB region be VB: V1 (OB region) = VB = G1 * f (c = 0) + O1 = O1.

Here, f (c) is a conversion function of the charge amount c into the output signal level, and G1 and G2 are AD-CH1 and AD.
Gain values in each channel of -CH2, O1 and O2 are offset values in each channel of AD-CH1 and AD-CH2, and the signal of each channel is represented by linear processing including gain processing and offset processing. Also, two types of predetermined charge amounts as pilot charges: c
When 1 and c2 are stored, the average value level of the signals in the detection region of these charges in AD-CH1 is V11,
Letting V21 and V22 be the average value levels of the signals in the detection regions of these charges in V12 and AD-CH2, respectively: V11 = G1 * f (c1) + O1, V12 = G1 * f (c2) + O1 V21 = G2 * f (c1) + O2, V22 = G2 * f (c2) + O2. From these equations, the relationship between the gain value and the offset value between the channels is expressed as follows.

G1 / G2 = (V11-V12) / (V21-V22) O2 = ((V21-V22) * O1 + (V11 * V22-V21 * V12)) / (V11
-V12) From this, in order to match the signal level of AD-CH2 with the signal level of AD-CH1, AD is adjusted with respect to the gain correction value GA1 and the offset correction value OF1 of AD-CH1.
-CH2 gain correction value GA2 and offset correction value O
F2 is GA2 = (G1 / G2) * GA1 = ((V11-V12) / (V21-V2
2)) * GA1 OF2 = (O1-O2) + OF1 = ((V11-V12-V21 + V22) * O1)-(V11 * V22-V21 * V12)) / (V11-V12) + OF1 = ((( V11-V12-V21 + V22) * VB)-(V11 * V22-V21 * V12)) / (V11-V12) + OF1.

Next, the case of FIG. 15B is shown in FIG.
The differences will be mainly described. In the case of FIG. 15B,
The difference between channels is detected using the signal of the pilot charge region outside the OB region which is different from the OB region which is an optical black level region and the image pickup region of the image sensor. Next, the detection and correction operation by the OB region and pilot charge in each channel will be described.

In the average value detection circuits 203 and 204, the OB shown in FIG. 15B set by the detection area setting circuit 209 is set.
The average value of the signals corresponding to the black level is detected using the signals in the upper area or the OB lower area (hatched area in the figure), and the pilot charge detection circuits 210 and 211 use the detection area setting circuit 20.
A signal corresponding to the pilot charge having a predetermined charge amount is detected using the signal of the pilot charge region (filled region in the drawing) shown in FIG. Using the detected data, the overall control circuit 121 shown in FIG. 13 uses the offset correction values (OF1, OF2) and the gain correction value (GA) which are correction data of the difference between channels.
1, GA2) to calculate the offset correction circuit 107, 1
The offset error is corrected at 08, and the signal level gain is corrected at the gain correction circuits 118 and 119.

The processing flow in the case of FIG. 15 (b) is shown in FIG. As shown in FIG. 18, first, the detection area setting circuit 209 sets the detection area (hatched area and filled area in the figure) shown in FIG. 15B as the detection area (process 60).
1) and then AD-CH1 in these detection areas
And the AD-CH2 signal from the average value detection circuit 203,
In 204, the average value of the signal corresponding to the black level is detected (process 608) using the signal in the OB upper region or the OB lower region (hatched region in the figure), and the pilot charge detection circuit 21
At 0 and 211, the average value of the signals in the detection region of the predetermined charge level is detected (process 610). Using this detection data, the overall control circuit 121 calculates an offset correction value corresponding to the black level correction (process 605), and calculates a gain correction value corresponding to the signal level correction (process 614).
To do.

Next, an example of the method of calculating the correction value will be described focusing on the points different from FIG. Figure 15
Similar to (a), AD-CH1 and AD-CH for a predetermined charge amount: c in the pilot charge corresponding image pickup device 116.
2 signals are respectively V1 = G1 * f (c) + O1 V2 = G2 * f (c) + O2, and when a predetermined amount of pilot charge: c1 is stored, this charge of AD-CH1 is When the average value level of the signal in the detection area is V11 and the average value level of the signal in the detection area of this charge in AD-CH2 is V21, V11 = G1 * f (c1) + O1 V21 = G2 * f (c1) + It becomes O2. Further, letting VB1 and VB2 be the average values detected from the signals corresponding to the black level using the signals in the OB region in AD-CH1 and AD-CH2, V1 (OB region) = VB1 = G1 * f (c = 0) + O1 = O1 V2 (OB area) = VB2 = G2 * f (c = 0) + O1 = O2, and the relationship between the gain value and the offset value between the channels is expressed by these equations.

G1 / G2 = (V11-VB1) / (V21-VB2) O1 = (VB1-VB2) + O2 From this, in order to match the signal level of AD-CH2 with the signal level of AD-CH1, AD AD for -CH1 gain correction value GA1 and offset correction value OF1
-CH2 gain correction value GA2 and offset correction value O
F2 is GA2 = (G1 / G2) * GA1 = (V11-VB1) / (V21-VB2) *
GA1 OF2 = (VB1-VB2) + OF1.

As described above, in FIG. 15B, the OB area and the pilot charge area, which are optical black level areas, are provided in each area, so that offset correction for black level detection and black level correction can be performed. It is possible to perform with high accuracy, and it is sufficient to store one charge amount as pilot charge, and it is possible to simplify the configuration of pilot charge of the image sensor.

Next, the case of FIG. 16A is shown in FIG.
The differences will be mainly described. In the case of FIG. 16 (a),
The black level is detected using the signal of the OB area on the left screen side, and the pilot charge area located in an area different from the imaging area of the image sensor is outside the imaging area shown in FIG. 15A (first pilot area). The difference between channels is detected using signals between the charge region) and the HCCD left screen and between the HCCD right screen and the output amplifier (second pilot charge region). next,
The detection and correction operation by the first pilot charge and the second pilot charge in each channel will be described.

The average value detection circuit 203 detects the average value of the signals corresponding to the black level by using the signals in the OB area (hatched area in the figure) set by the detection area setting circuit 209 and shown in FIG. The pilot charge detection circuits 210 and 211 use a signal of the second pilot charge region (filled region in the drawing) shown in FIG. 16A set by the detection region setting circuit 209 to generate a pilot charge having a predetermined charge amount. A corresponding signal is detected, and a signal in the first pilot charge region (filled region in the figure) is used to detect a signal corresponding to a pilot charge having a predetermined charge amount. Using the detected data, the overall control circuit 121 shown in FIG. 13 uses offset correction values (OF1, OF2) and gain correction values (G
A1, GA2) is calculated, and the offset correction circuit 107,
The offset error is corrected at 108, and the signal level gain is corrected at the gain correction circuits 118 and 119.

The processing flow in the case of FIG. 16 (a) is shown in FIG. As shown in FIG. 19, first, the detection area setting circuit 209 sets the OB area and the pilot charge area (hatched area and two filled areas in the figure) shown in FIG. 16A as the detection area (process 601). . Next, the average value detection circuit 203 detects the average value of the signal corresponding to the black level from the signal in the OB area (in this case, output from AD-CH1) (step 608), and AD-CH1 and AD- in the pilot charge area. From the CH2 signal, the pilot charge detection circuits 210 and 211 detect the first pilot charge region (process 615) and the second pilot charge region (process 617), and calculate the average value of the signals of the first pilot charge region. Detection (process 616) and detection of the average value of the signals in the second pilot charge region (process 618) are performed. Using this detection data, the overall control circuit 121 calculates an offset correction value corresponding to the black level correction and a gain correction value corresponding to the signal level correction (process 61).
3) Do.

Next, an example of the method of calculating the correction value will be described. Signals of AD-CH1 and AD-CH2 corresponding to a predetermined charge amount: c in the second pilot charge region in the pilot charge corresponding image pickup device 116 are respectively V1 = (FG1 * g (c)) * SG1 + O1 V2 = ( FG2 * g (c)) * SG2 + O2 Also, AD for a predetermined charge amount: c in the first pilot charge region in the pilot charge corresponding image sensor 116.
-CH1 and AD-CH2 signals are respectively V1 = (FG1 * (h (c) + HV1)) * SG1 + O1 = (FG1 * SG1) * h (c) + (FG1 * SG1 * HV1 + O1) V2 = (FG2 * (h (c) + HV2)) * SG2 + O2 = (FG2 * SG2) * h (c) + (FG2 * SG2 * HV2 + O2) and the output signal of the OB area is VB Then, V1 (OB area) = VB = (FG1 * (h (c = 0) + HV1)) * SG1 + O1 = FG1 * HV1 * SG1 + O1.

Here, g (c) is a conversion function of the charge amount c in the second pilot charge region into the output signal level,
FG1 and FG2 are gain values of the output amplifier in the image sensor in each channel of AD-CH1 and AD-CH2, S
G1 and SG2 are gain values after the image sensor output in each channel of AD-CH1 and AD-CH2, and O1 and O2 are offset values after the image sensor output in each channel of AD-CH1 and AD-CH2. Also, h (c)
Is a conversion function of the charge amount c in the first pilot charge region into an output signal level, and HV1 and HV2 are AD-CH.
1 and V-direction CC in each channel of AD-CH2
It is an offset value in the image pickup device such as CCD in the D and H directions.

When two kinds of predetermined charge amounts: c21 and c22 are stored as the charges in the second pilot charge region,
Letting V121 and V122 be the average value levels of these charge detection regions in AD-CH1, respectively, and let V221 and V222 be the average value levels of these charge detection region signals in AD-CH2, then V121 = (FG1 * g (c21)) * SG1 + O1 V122 = (FG1 * g (c22)) * SG1 + O1 V221 = (FG2 * g (c21)) * SG2 + O2 V222 = (FG2 * g (c22)) * SG2 + O2 Further, when two kinds of predetermined charge amounts: c11 and c12 are stored as the charges in the first pilot charge region, AD-CH1
The average value levels of the signals in these charge detection regions are V111 and V112, and the average value levels of the signals in these charge detection regions in AD-CH2 are V22, respectively.
1, V212: V111 = (FG1 * (h (c11) + HV1)) * SG1 + O1 V112 = (FG1 * (h (c12) + HV1)) * SG1 + O1 V211 = (FG2 * (h ( c11) + HV2)) * SG2 + O2 V212 = (FG2 * (h (c12) + HV2)) * SG2 + O2, and the relationship between the gain value and the offset value between channels is shown in the table below. To be done.

(FG1 * SG1) / (FG2 * SG2) = (V121-V
122) / (V221-V222) (FG1 * SG1 * HV1 + O1)-(FG2 * SG2 * HV2 + O2) =
((V111 * V221-V111 * V222-V211 * V121 + V211 * V122)
-(V221-V222-V121 + V122) * VB) / (V121-V122) Therefore, in order to match the signal level of AD-CH2 with the signal level of AD-CH1, the gain correction value GA1 of AD-CH1 and AD for offset correction value OF1
-CH2 gain correction value GA2 and offset correction value O
F2 is GA2 = ((FG1 * SG1) / (FG2 * SG2)) * GA1 = ((V121-V122) / (V221-V222)) * GA1 OF2 = ((FG1 * SG1 * HV1 + O1)-( FG2 * SG2 * HV2 + O2)) + OF1 = (((V111 * V221-V111 * V222-V211 * V121 + V211 * V122)-(V221-V222-V121 + V122) * VB) / (V121-V122) ) + OF1.

As described above, in FIG. 16A, by having two pilot charge regions, the factors that cause a difference between a plurality of output signals are the gain difference and offset difference in the image pickup device and the output after the image pickup device. The difference between the channels can be detected with high accuracy by using the gain difference and the offset difference in (1), and the difference due to the factors can be corrected.

Next, the case of FIG. 16B is shown in FIG.
The differences will be mainly described. In the case of FIG. 16 (b),
The difference between channels is detected by using the signal of the pilot charge region located in a region different from the OB region which is an optical black level region and the image capturing region of the image sensor shown in FIG. 16A. Next, the OB area in each channel and the first
The detection and correction operations using the pilot charge and the second pilot charge will be described.

In the average value detection circuits 203 and 204, the OB shown in FIG. 16B set by the detection area setting circuit 209 is set.
The pilot charge detection circuit 21 detects the average value of the signals corresponding to the black level by using the signals in the upper area (hatched area in the figure).
In Nos. 0 and 211, a signal corresponding to a predetermined amount of pilot charge is generated by using the signal of the second pilot charge region (filled region in the drawing) shown in FIG. 16B set by the detection region setting circuit 209. In addition, the signal corresponding to the pilot charge having a predetermined charge amount is detected by using the signal of the first pilot charge region (filled region in the drawing). The offset correction values (OF1, OF2) and the gain correction values (GA1, GA2), which are the correction data of the difference between the channels, are calculated in the overall control circuit 121 shown in FIG. , 108 corrects the offset error,
The gain correction circuits 118 and 119 perform signal level gain correction.

FIG. 20 shows a processing flow in the case of FIG. 16 (b). First, the detection area setting circuit 209 sets the detection areas (hatched areas and filled areas in the figure) shown in FIG. 16B as the detection areas (process 601), and then AD-CH1 and AD in these detection areas. From the -CH2 signal, the average value detection circuits 203 and 204 detect the average value of the signal corresponding to the black level by using the signal in the OB upper area (hatched area in the figure) (process 608), and the pilot charge detection circuit At 210 and 211, the first pilot charge region is detected (process 615) or the second pilot charge region is detected (process 617), and the average value of the signals in this first pilot charge region is detected (process 616) or the second value. The average value of the signals in the pilot charge region is detected (process 618). Using this detection data, the overall control circuit 121 calculates an offset correction value corresponding to the black level correction (process 605), and calculates a gain correction value corresponding to the signal level correction (process 614).

Next, a method of calculating this correction value will be described. AD for a predetermined charge amount: c in the second pilot charge region in the pilot charge corresponding imaging device 116
The signals of -CH1 and AD-CH2 are V1 = (FG1 * g (c)) * SG1 + O1 V2 = (FG2 * g (c)) * SG2 + O2, respectively. AD for a predetermined charge amount: c in one pilot charge region
-CH1 and AD-CH2 signals are respectively V1 = (FG1 * (h (c) + HV1)) * SG1 + O1 = (FG1 * SG1) * h (c) + (FG1 * SG1 * HV1 + O1) V2 = (FG2 * (h (c) + HV2)) * SG2 + O2 = (FG2 * SG2) * h (c) + (FG2 * SG2 * HV2 + O2), and AD-CH1 and AD-CH2 At O
Letting VB1 and VB2 be the average values detected from the signal corresponding to the black level using the signal in the B area, V1 (OB area) = VB1 = (FG1 * (h (c = 0) + HV1)) * SG1 + O1 = FG1 * HV1 * SG1 + O1 V2 (OB area) = VB2 = (FG2 * (h (c = 0) + HV2)) * SG2 + O2 = FG2 * HV2 * SG2 + O2.

Here, g (c) is a conversion function for converting the charge amount c in the second pilot charge region into an output signal level, and FG1 and FG2 are output amplifiers in the image pickup device in each channel of AD-CH1 and AD-CH2. , SG1 and SG2 are gain values after the image sensor output in each channel of AD-CH1 and AD-CH2, O
1 and O2 are offset values after the image sensor output in each channel of AD-CH1 and AD-CH2. Further, h (c) is the charge amount c in the first pilot charge region.
The output signal level conversion functions HV1 and HV2 are the offset values in the image sensor of the V-direction CCD and the H-direction CCD in the AD-CH1 and AD-CH2 channels.

Next, the pilot charge region in this case will be described. In the first example, when two kinds of predetermined charge amounts: c21 and c22 are stored as the charges in the second pilot charge region, the average value levels of the signals in the detection region of these charges in AD-CH1 are V121 and V122, respectively. , AD-CH
Letting V221 and V222 be the average value levels of the signals in the detection region of these charges in 2 respectively, V121 = (FG1 * g (c21)) * SG1 + O1 V122 = (FG1 * g (c22)) * SG1 + O1 V221 = (FG2 * g (c21)) * SG2 + O2 V222 = (FG2 * g (c22)) * SG2 + O2, and the relationship between gain value and offset value between channels is shown in the following formula. To be done.

(FG1 * SG1) / (FG2 * SG2) = (V121-V
122) / (V221-V222) (FG1 * SG1 * HV1 + O1)-(FG2 * SG2 * HV2 + O2) =
VB1-VB2 From this, in order to match the signal level of AD-CH2 with the signal level of AD-CH1, AD is adjusted with respect to the gain correction value GA1 and the offset correction value OF1 of AD-CH1.
-CH2 gain correction value GA2 and offset correction value O
F2 is GA2 = ((FG1 * SG1) / (FG2 * SG2)) * GA1 = ((V121-V122) / (V221-V222)) * GA1 OF2 = ((FG1 * SG1 * HV1 + O1)-( FG2 * SG2 * HV2 + O2)) + OF1 = (VB1-VB2) + OF1.

In the second example, when two kinds of predetermined charge amounts: c11 and c12 are stored as the charges in the first pilot charge region, the average value levels of the signals in the detection regions of these charges in AD-CH1 are respectively set. V111, V112, AD-CH
Letting V221 and V212 be the average value levels of these charge detection regions in 2 respectively, V111 = (FG1 * (h (c11) + HV1)) * SG1 + O1 V112 = (FG1 * (h (c12) + HV1)) * SG1 + O1 V211 = (FG2 * (h (c11) + HV2)) * SG2 + O2 V212 = (FG2 * (h (c12) + HV2)) * SG2 + O2 From these formulas The relationship between the gain value and the offset value between channels is expressed as follows.

(FG1 * SG1) / (FG2 * SG2) = (V111-V
112) / (V211-V212) (FG1 * SG1 * HV1 + O1)-(FG2 * SG2 * HV2 + O2) =
VB1-VB2 From this, in order to match the signal level of AD-CH2 with the signal level of AD-CH1, AD is adjusted with respect to the gain correction value GA1 and the offset correction value OF1 of AD-CH1.
-CH2 gain correction value GA2 and offset correction value O
F2 is GA2 = ((FG1 * SG1) / (FG2 * SG2)) * GA1 = ((V111-V112) / (V211-V212)) * GA1 OF2 = ((FG1 * SG1 * HV1 + O1)-( FG2 * SG2 * HV2 + O2)) + OF1 = (VB1-VB2) + OF1.

As described above, in FIG. 16B, the black level is detected and the black level is corrected by having two levels of charge in the OB region and the pilot charge region, which are optical black level regions in each region. It is possible to perform the offset correction with high accuracy, and the factors that cause the difference between the plurality of output signals are the gain difference and the offset difference within the image sensor and the gain difference and the offset after the image sensor output. By using the difference, the difference between the channels can be detected with high accuracy, and the difference due to the factor can be corrected.

As described above, in the fourth embodiment,
By having the OB region and the pilot charge region and detecting the average value data in each detection region, the error between channels can be detected with high accuracy without being affected by the subject, and the error between channels can be corrected. It is possible.

(Fifth Embodiment) FIG. 21 is a block diagram showing an example of the hardware configuration of an image pickup apparatus according to the fifth embodiment of the present invention. Hereinafter, differences from the fourth embodiment will be mainly described.

Portions corresponding to those of the structure of the fourth embodiment shown in FIG. 13 are designated by the same reference numerals. What is different is a pilot charge correction circuit 123 which is a charge signal correction means and an inter-channel detection circuit 124 corresponding to the pilot charge correction circuit 123. In FIG. 21, 116 is a plurality of (here, two) imaging regions and a predetermined number. Corresponding to a pilot charge such as a CCD, which is an image pickup unit that independently outputs (CH1 and CH2) signals of an area for storing the electric charge (hereinafter, also referred to as pilot charge), a plurality of these imaging areas, and an area for storing a predetermined electric charge. An element 117 is a timing signal generation circuit for generating a timing signal necessary for driving and signal processing of the pilot charge corresponding imaging element 116, 103, 10
Reference numeral 4 is an analog signal processing circuit for performing noise removal, gain control, etc., and 105 and 106 are analog-to-digital conversion circuits (hereinafter referred to as A / D) for converting an analog signal into a digital signal.
(Referred to as conversion circuits), 107 and 108 are correction means for correcting the difference between channels of two signals, and an offset correction circuit for correcting the offset component of the input signal, and 118 and 119 are correction means for correcting the difference between channels of two signals. A gain correction circuit that corrects the gain of the input signal, 111 is an image composition circuit that combines a plurality of video signals, 112 is a camera signal processing circuit, 123 is a pilot charge correction circuit that corrects the pilot charges, and 124 is a difference between the plurality of signals. Is a channel-to-channel detection circuit for detecting the signal, 125 is an overall control circuit for controlling the circuit, and 122 is a mode setting circuit.

The operation of the image pickup apparatus configured as described above will be described below.

In FIG. 21, the pilot charge-corresponding image pickup device 116 divides the subject image into a plurality of image pickup regions by the drive signal from the timing signal generating circuit 117, picks up an image, and outputs two signals (from each of the plurality of image pickup regions). CH1 and CH2) are also output, and the signals in the pilot charge region are also output as two signals (CH1 and CH2) corresponding to a plurality of imaging regions. The data output from the right half surface of the image sensor 116 corresponding to pilot charge is
The signal is input to the analog signal processing circuit 104 via the CH1 output, and a known method such as correlated double sampling is performed here to remove reset noise and the like contained in the output of the CCD and the like, and to a predetermined signal level. Activate the AGC circuit for amplifying the output. The output after this AGC is A
By inputting to the / D conversion circuit 106, it is converted into a digital signal and an output of AD-CH1 is obtained.

Similarly, the data output from the left half surface of the image sensor is sent to the analog signal processing circuit 10 via the CH2 output.
3 and the same method such as correlated double sampling is performed here to remove the reset noise and the like contained in the output of the CCD and the like, and an AGC circuit for amplifying the output to a predetermined signal level. To work. By inputting the output after this AGC to the A / D conversion circuit 105, it is converted into a digital signal and an output of AD-CH2 is obtained.

Next, regarding the output signals of AD-CH1 and AD-CH2, first, the output signal corresponding to the pilot charge in the image pickup device is a pilot charge correction unit which is a detection unit and a correction unit of variation when the pilot charge is injected. Circuit 123
The signal corresponding to the corrected pilot charge and the signal corresponding to the image pickup area from the image pickup element are simultaneously input to the inter-channel detection circuit 124 which is a detection means for detecting a difference between the channels. The detection data detected by the inter-channel detection circuit 124 is input to the overall control circuit 125.

The output signals of AD-CH1 and AD-CH2 are offset correction circuits 107 and 108 which are correction means for correcting the difference between the channels of both output signals.
Is input to these offset correction circuits 107 and 108.
Then, the predetermined offset correction control signal OF1, which is calculated and set by the overall control circuit 125, which is the other input signal.
Offset correction processing of both signals is performed using OF2. Next, the output signals of the offset correction circuits 107 and 108 are
The signals are input to gain correction circuits 118 and 119 which are another correction means for correcting the difference between the channels of both output signals. In these gain correction circuits 118 and 119, the overall control circuit 125 which is the other input signal. The gain correction processing of both signals is performed using the predetermined gain correction control signals GA1 and GA2 calculated and set in step 1.

As described above, the two output signals are detected by the inter-channel detection circuit by using the signal corresponding to the pilot charge corrected by the pilot charge correction circuit 123 and converted into the control signal calculated by the overall control circuit 125. The difference between channels is corrected by an offset correction circuit and a gain correction circuit which are correction means for performing correction based on the above.
Here, the correction of the difference between the channels may be performed only by the offset correction or the gain correction depending on the case.

In this way, the image data output in which the difference between the channels generated between the two outputs is corrected is converted into one image data by the image synthesizing circuit 111 (the left and right outputs are made into one output), and Stage camera signal processing circuit 11
In step 2, predetermined signal processing (color interpolation processing, γ conversion, etc.) is performed.

Next, variations in pilot charges will be described with reference to FIG. In FIG. 22, FIG. 22 shows the pilot charge corresponding image pickup device 116 having the pilot charge region shown in FIG. 21, and the charge level is not uniform in the pilot charge region generated when the pilot charge is injected from the division boundary surface. Shows the case. Next, FIG. 23 and FIG. 24 show an example of a state where the pilot charge level generated in the pilot charge region is not uniform.

FIG. 23 shows a case where the pilot charge level fluctuates in pixel units as the charge is injected from the vicinity of the division boundary surface as it advances in the left and right ends. Figure 24
Shows the case where the pilot charge level is lowered or increased as the charge is injected from the vicinity of the division boundary surface in the left and right ends. At this time, the variation of the pilot charge level is often substantially bilaterally symmetric, but it is not limited to this. In this way, the level of the injected charge may become non-uniform in the pilot charge region when the pilot charge is injected.

Next, correction for such nonuniformity of pilot charge levels will be described. A concrete configuration of the pilot charge correction circuit 123, which is a detection unit and a correction unit of the variation at the time of injecting the pilot charge shown in FIG. 21, will be described with reference to the block diagram of the configuration example shown in FIG.

In FIG. 25, 401 is a correction area setting circuit for setting a correction area, 402 and 403 are pixel fluctuation detection circuits for input signals corresponding to pilot charges in the set range, and 404 and 405 are in the set range. A histogram detection circuit for the input signal corresponding to the pilot charge, 406 is a correction control circuit that generates correction data for correcting the signal corresponding to the pilot charge using the data from each detection circuit, and 407 and 408 are correction means. An averaging correction circuit that performs averaging processing of a certain input signal,
Reference numerals 409 and 410 are gain correction circuits that are correction means for correcting the signal level of the input signal. Hereinafter, an example of the operation of the pilot charge correction circuit 123 configured as above will be described with reference to FIGS. 26 and 27.

In FIG. 25, first, the correction area setting circuit 401 sets the area for correcting the signal corresponding to the pilot charge. This correction area setting circuit 401 is shown in FIG.
VD / from the timing signal generation circuit 117 shown in FIG.
Image sensor 1 corresponding to pilot charge based on HD signal
The correction range for the output signal from 16 is determined and the timing of detection by each detection circuit is set. First, in the pixel variation detection circuits 402 and 403 in each channel, as shown in FIG. 26, for example, charges of the same level are injected from the division boundary surface toward the left and right ends in the correction area set by the correction area setting circuit 401. The left screen averaging process and the right screen averaging process are performed in the area, and the signal level after averaging within the set averaging process range and the signal level of each pixel are used to calculate the pilot charge in pixel units. To detect the fluctuation amount of. Further, in the histogram detection circuits 404 and 405 for the input signal corresponding to the pilot charge, as shown in FIG. 27, the pilot detection is performed in the correction area set by the correction area setting circuit 401, for example, toward the left and right ends from the division boundary surface. Histogram data is detected on the left screen and the right screen in the region where the charge level is changed from the maximum level to the minimum level and injected.

The histogram data detection at this time will be described. As shown in FIG. 27A, when variations such as variations do not occur in the injection of pilot charges, the histogram data of the signal corresponding to the pilot charges has the same value for all gray levels of both channels. . On the other hand, when variations such as variations occur in the injection of pilot charge as shown in FIG. 27B, the histogram data of the signal corresponding to the pilot charge is the value changed in the gradation of each channel. Become. In this way, it is possible to detect the variation amount with respect to the injected pilot charge from the histogram data of the signal corresponding to the pilot charge level.

Next, using these detected data, as shown in FIG.
In the correction control circuit 406 shown in FIG. 5, correction data (AVE1, AVE2, GA
1, GA2) is obtained. An example of a method of calculating the correction data and a correction method using the correction data will be described.

First, the averaging processing in the averaging correction circuits 407 and 408 is set according to the variation amount in pixel units detected in the pixel variation detection circuits 402 and 403. This is because, for example, the fluctuation of each pixel generated in the region where the charges of the same level are injected shown in FIG. 23 and the signal level at the pilot charge region position generated in the region where the charges of the same level shown in FIG. 24 are generated. The averaging processing range (number of pixels) that performs optimal averaging with respect to changes in decrease or increase is calculated, and the calculated correction data (AVE1, A
VE2) is used to perform averaging processing in averaging correction circuits 407 and 408. Next, the histogram detection circuits 404 and 40
From the fluctuation amount of the histogram data detected in 5, the input signal level with a small fluctuation amount is selected from the fluctuation amount with respect to the input signal level in the same channel and the fluctuation amount with respect to the input signal level between the channels, and the gain for the selected input signal level is selected. Correction value (GA1, GA2)
Is calculated, and the gain correction circuits 409 and 410 perform gain correction using the calculated correction data (GA1, GA2).

As described above, the pilot charge correction circuit 123
In the above, the pixel variation detection circuits 402 and 403 and the histogram detection circuits 404 and 405 can detect variations and variations in the charge level in the pilot charge injection, and the averaging correction circuits 407 and 408 and the gain correction circuits 409 and 410. It is possible to perform variation (variation) correction according to the variation amount.

As described above, in the fifth embodiment,
The pilot charge correction circuit 123 is provided. The pilot charge correction circuit 123 detects the variation and fluctuation amount of the signal corresponding to the pilot charge with high accuracy to detect the reliability of the pilot charge. For example, the reliability is low. In this case, it is possible to select a channel-to-channel correction method that does not use pilot charges. Further, it is possible to further correct the variation and fluctuation amount of the signal corresponding to the pilot charge, and to perform the highly accurate inter-channel correction using the pilot charge even when the variation occurs in the pilot charge injection.

The advantages of the first to fifth embodiments described in detail above can be summarized as follows.

(1) In the first embodiment of the present invention,
A histogram detection circuit, which is a signal level distribution detection means, is provided to detect error between channels with high accuracy by detecting the histogram data of the signals in the respective divided areas. It is possible to have a circuit, calculate a histogram correction value that matches the histogram data for each channel, and correct the error between channels in the histogram correction circuit. In addition, the reliability of the data from the image detection area can be determined by using the data from the image detection area, and the reliability of the error detection between channels can be improved by using the reliable data. Is.

(2) In the second embodiment of the present invention,
A histogram detection circuit, which is a signal level distribution detection means, is provided to detect error between channels with high accuracy by detecting the histogram data of the signals in the respective divided areas. It is possible to have a circuit, calculate a histogram correction value for matching the histogram data of each channel, and repeatedly correct the error between the channels in the histogram correction circuit to perform the accurate inter-channel correction.

(3) In the third embodiment of the present invention,
The signal level distribution detecting means has an average value detecting circuit for all channel signals and a histogram detecting circuit for all channel signals. It is possible to detect an error with high accuracy and to correct an error between channels with high accuracy.

(4) In the fourth embodiment of the present invention,
It has an OB region and a pilot charge region that is a region for storing a predetermined charge, and by detecting the average value data in each detection region, the error between channels can be detected with high accuracy without being affected by the subject. It is possible to correct the error between them.

(5) In the fifth embodiment of the present invention,
It has a pilot charge correction circuit which is a charge signal correction means, and in this pilot charge correction circuit, the variation of the signal corresponding to the pilot charge and the reliability of the pilot charge are detected by detecting the variation amount with high accuracy,
For example, if the reliability is low, it is possible to select the channel-to-channel correction that does not use the pilot charge.
Further, it is possible to correct the variation and fluctuation amount of the signal corresponding to the pilot charge, and to perform the highly accurate inter-channel correction using the pilot charge even when the variation occurs in the pilot charge injection.

In the third embodiment, as a setting example of the detection area setting circuit, a specific position (indicated by a dotted line in the figure) in the lateral direction from the vicinity of the division boundary of the image data of the imaging area shown in FIG. Although the case of setting up to the area) has been described, the imaging area of the image sensor shown in FIG. 6B includes an OB area (shaded area in the drawing) which is an optical black level area, and The same effect can be obtained even when a specific position (dotted line area in the drawing) is set in the lateral direction from the vicinity of the division boundary of the image data.

Further, in the fourth embodiment, the case where the image pickup device corresponding to the pilot charge is provided based on the circuit configuration shown in the first embodiment has been described, but it is shown in the second embodiment and the third embodiment. Even if the image pickup device corresponding to the pilot charge is provided based on the above circuit configuration, the same effect can be obtained.

Further, in the fifth embodiment, the case where the image pickup device corresponding to the pilot charge and the pilot charge correction circuit are provided based on the circuit configuration shown in the first embodiment has been described. Even when the image pickup device corresponding to the pilot charge and the pilot charge correction circuit are provided based on the circuit configuration shown in the third form, the same effect can be obtained.

Further, in the fifth embodiment, the case where the gain correction circuit is provided as the pilot charge correction circuit and the gain correction is performed for the input signal level having a small fluctuation amount has been described. On the other hand, the configuration may be such that each gain is corrected, or a histogram correction circuit that performs gain correction for each gradation may be provided.

Further, in the above-mentioned embodiment, the case where the image pickup device has the image pickup area divided into two has been described, but the present invention is not limited to this, and the image pickup is divided into, for example, three or four. The same effect can be obtained in the element. Further, in the above-described embodiment, the case where the output signals from the two divided image pickup areas are output from the boundary surface to both end directions has been described, but the present invention is not limited to this. Alternatively, the same effect can be obtained in the case of outputting in the same direction from each imaging region.

In the above embodiment, the signal level detected by the histogram detection circuit, the average value detection circuit, etc. is not particularly described, but the detection of this signal level is the same in each divided area. If the image sensor has a color filter and the output signal corresponding to this color filter is used for detection, the signal level of the luminance signal created by calculation from the output signal corresponding to this color filter Or detection using the signal level of the output signal corresponding to the same color filter. When the image pickup unit has a prism and the output signal corresponding to the light of each wavelength split by the prism is used for detection, a luminance signal created by calculation from the output signal corresponding to the light of each wavelength. Detection using the signal level of, or detection using the signal level of the output signal corresponding to the light of the same wavelength.

[0203]

As described above, the present invention has the following effects.

(1) It has a signal level distribution detecting means, detects the level distribution of a plurality of output signals from each divided image pickup area, and detects an error between the plurality of output signals.
Further, since a division signal correction means is provided and an error between the plurality of output signals is corrected so that the signal level distributions between the plurality of output signals are made to coincide with each other, the subject with non-linear element variations such as changes in contrast. Even when shooting
It is possible to highly accurately correct the imbalance between a plurality of output signals.

(2) Further, with respect to factors such as a signal in a saturated region and a signal in a boundary region of different subjects, which reduces reliability, reliability detection means is provided, and signal levels of a plurality of output signals are provided. Is within a certain range, and
Reliability is judged based on whether the signal level difference between multiple output signals is within a certain range, signals in unreliable areas are removed, and multiple signals are detected using the detection signal when there is reliability. By performing correction between the output signals of, it is possible to perform more accurate correction.

(3) Further, a region (pilot charge region) for storing a predetermined charge in the image pickup unit is provided separately from the image detection region, and a signal between a plurality of output signals obtained without being influenced by the subject is provided. If a plurality of output signals from the image detection area are corrected based on the above, more accurate correction becomes possible.

(4) Furthermore, for correction accuracy deterioration due to variations in pilot charges in the pilot charge region, charge signal correction means is added to correct the variations in pilot charge, and pilot charges are used. The accuracy of the correction can be made higher.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an image pickup apparatus according to a first embodiment of the present invention.

FIG. 2 is a block diagram of an inter-channel detection circuit in the image pickup apparatus according to the first embodiment.

FIG. 3 is a block diagram of a divided signal correction circuit in the image pickup apparatus according to the first embodiment.

FIG. 4 is an explanatory diagram of histogram correction in a histogram correction circuit in the image pickup apparatus according to the first embodiment (No. 1).

FIG. 5 is an explanatory diagram of histogram correction in the histogram correction circuit in the image pickup apparatus according to the first embodiment (No. 2).

FIG. 6 is a setting example of a detection area setting circuit as a specific description of correction between channels in the image pickup apparatus according to the first embodiment.

FIG. 7 is a signal processing flowchart (part 1) in the setting example of the detection area setting circuit in the image pickup apparatus according to the first embodiment.

FIG. 8 is a signal processing flowchart (part 2) in the setting example of the detection area setting circuit in the image pickup apparatus according to the first embodiment.

FIG. 9 is a signal processing flowchart (part 3) in the setting example of the detection area setting circuit in the image pickup apparatus according to the first embodiment.

FIG. 10 is a block diagram showing a configuration of an image pickup apparatus according to a second embodiment of the present invention.

FIG. 11 is a block diagram of an inter-channel detection circuit of an image pickup device according to a third embodiment of the present invention.

FIG. 12 is a signal processing flow chart in a setting example of a detection area setting circuit in the image pickup apparatus according to the third embodiment.

FIG. 13 is a block diagram showing a configuration of an image pickup apparatus according to a fourth embodiment of the present invention.

FIG. 14 is a block diagram of an inter-channel detection circuit in the image pickup device according to the fourth embodiment.

FIG. 15 is a setting example of a detection area setting circuit (part 1) as a specific description of correction between channels in the image pickup apparatus according to the fourth embodiment.

FIG. 16 is a setting example of a detection area setting circuit (part 2) as a specific description of correction between channels in the image pickup apparatus according to the fourth embodiment.

FIG. 17 is a signal processing flowchart (1) in a setting example of the detection area setting circuit in the image pickup apparatus according to the fourth embodiment.

FIG. 18 is a signal processing flowchart (part 2) in the setting example of the detection area setting circuit in the imaging apparatus according to the fourth embodiment.

FIG. 19 is a signal processing flowchart (3) in the setting example of the detection area setting circuit in the image pickup apparatus according to the fourth embodiment.

FIG. 20 is a signal processing flowchart (4) in the setting example of the detection area setting circuit in the image pickup apparatus according to the fourth embodiment.

FIG. 21 is a block diagram showing a configuration of an image pickup device according to a fifth embodiment of the present invention.

FIG. 22 is an explanatory diagram of a pilot charge in the image pickup apparatus according to the fifth embodiment (part 1).

FIG. 23 is an explanatory diagram of a pilot charge in the image pickup apparatus according to the fifth embodiment (part 2).

FIG. 24 is an explanatory view (No. 3) of pilot charges in the image pickup apparatus according to the fifth embodiment.

FIG. 25 is a block diagram of a pilot charge correction circuit in the image pickup apparatus according to the fifth embodiment.

FIG. 26 is an explanatory diagram of averaging processing operation of the pilot charge correction circuit in the image pickup apparatus according to the fifth embodiment.

FIG. 27 is an explanatory diagram of a histogram processing operation of the pilot charge correction circuit in the image pickup apparatus according to the fifth embodiment.

[Explanation of symbols]

101 Imaging unit 102, 117 timing signal generation circuit 103, 104 analog signal processing circuit 105, 106 analog-digital conversion circuit 107, 108 Offset correction circuit 109, 110 Histogram correction circuit 111 Image synthesis circuit 112 Camera signal processing circuit 113, 120, 124 inter-channel detection circuit 114, 121, 125 Overall control circuit 115, 122 mode setting circuit 116 Pilot charge compatible image sensor 118,119 Gain correction circuit 123 Pilot charge correction circuit 130, 131 division signal correction circuit 201,209 detection area setting circuit 202 Reliability area detection circuit 203, 204 Averaging correction circuit 205, 206 Histogram detection circuit 207 All channel average value detection circuit 208 All channel histogram detection circuit 210,211 Pilot charge detection circuit 230,231 Peak detection circuit 301, 302 offset correction circuit 303, 304 gain correction circuit 305,306 Histogram correction circuit 401 correction area setting circuit 402,403 Pixel change detection circuit 404,405 Histogram detection circuit 406 Correction control circuit 407, 408 Averaging correction circuit 409,410 Gain correction circuit

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Bunki Shibuya             1006 Kadoma, Kadoma-shi, Osaka Matsushita Electric             Sangyo Co., Ltd. F-term (reference) 5C024 CX27 HX18 HX20 HX21 HX29

Claims (23)

[Claims] 1. An image pickup unit that divides a subject image into a plurality of image pickup regions and picks up an image, and outputs a signal from each of the plurality of image pickup regions; and a plurality of output signals output from each of the plurality of image pickup regions. An image pickup apparatus comprising: the signal level distribution detection means of claim 1; and divided signal correction means for correcting the plurality of output signals based on the detection result of the signal level distribution detection means. 2. The signal level distribution detection means detects the signal level distribution of each output signal of the plurality of output signals output from each of the plurality of imaging regions and the signal level distribution of all output signals. The image pickup apparatus according to claim 1, wherein the image pickup apparatus is provided. 3. The signal level distribution detecting means includes detection of a signal level of a black area in a plurality of output signals output from each of the plurality of imaging areas. The imaging device according to 2. 4. An image pickup unit which divides a subject image into a plurality of image pickup areas and picks up an image, and outputs a signal from each of the plurality of image pickup areas, and a plurality of output signals output from each of the plurality of image pickup areas. A reliability detecting means for determining reliability of the second detecting means with respect to the plurality of output signals based on a detection result of the first detecting means, and the reliability detecting means is reliable. And a divided signal correction unit that corrects the plurality of output signals based on the detection result of the second detection unit that is determined to have the characteristic. 5. The detection result of the second detection means, which is determined by the reliability detection means to be reliable, is the signal level of the black area in the plurality of output signals output from each of the plurality of imaging areas. It is a detection result, It is characterized by the above-mentioned.
The imaging device according to. 6. The detection result of the second detecting means that the reliability detecting means determines to be reliable is the detection result of the second detecting means immediately before the reliability detecting means determines to be reliable. The imaging device according to claim 4, wherein the imaging result is obtained. 7. The first detecting means and the second detecting means are any one of a signal level distribution detecting means, a signal average value detecting means, and a signal peak detecting means. The image pickup apparatus according to claim 6. 8. The method according to claim 4, wherein the first detecting means is a signal level distribution detecting means, and the second detecting means is a signal average value detecting means or a signal peak detecting means. The imaging device according to any one of 6 above. 9. An imaging unit, which has an area for storing a predetermined charge, divides a subject image into a plurality of imaging areas, and outputs a signal from each of the plurality of imaging areas; An image pickup apparatus comprising: a signal level distribution detecting unit for a plurality of output signals output from each of them; and a split signal correcting unit for correcting the plurality of output signals based on a detection result of the signal level distribution detecting unit. 10. The signal level distribution detection means detects a signal level distribution of a plurality of output signals output from a region in the plurality of imaging regions where predetermined charges are stored. Imaging device. 11. The image pickup apparatus according to claim 9, wherein the signal level distribution detection means detects a signal level distribution of a plurality of output signals output from a plurality of image pickup areas corresponding to a subject image. . 12. The signal level distribution detecting means includes detection of signal levels of a plurality of output signals output from optical black level regions in the plurality of image pickup regions. Item 12. The imaging device according to any one of items 11. 13. The image pickup apparatus according to claim 9, wherein the predetermined electric charge is injected from the outside of the image pickup device. 14. The image pickup device according to claim 9, wherein the predetermined charges are injected from outside the image pickup device at a charge level in a desired range. 15. The divided signal correction means includes correction data calculation means for calculating a correction amount for correcting a plurality of output signals based on a detection result of the signal level distribution detection means. The imaging device according to any one of claims 1 to 3 or 9 to 14. 16. The divided signal correction means calculates a correction amount for correcting a plurality of output signals based on a detection result of the signal level distribution detection means, and the correction data calculation means. 4. At least one of an offset correction means for performing an offset correction for the plurality of output signals based on an output and a histogram correction means for performing a histogram correction, respectively. Item 15. The imaging device according to any one of items 14. 17. The image pickup apparatus according to claim 16, wherein the histogram correction unit equalizes the appearance frequencies for each gradation with respect to the plurality of output signals. 18. An imaging unit which has an area for storing a predetermined electric charge, divides a subject image into a plurality of imaging areas, picks up an image, and outputs a signal from each of the plurality of imaging areas, and stores the predetermined electric charge. A charge signal correction unit that corrects the plurality of output signals output from the region, a signal level detection unit for the signal corrected by the charge signal correction unit, and a plurality of the plurality of units based on the detection result of the signal level detection unit. An image pickup apparatus comprising: a split signal correction unit that corrects an output signal. 19. The charge signal correction means has a pixel change detection circuit for detecting a change amount in pixel units of the plurality of output signals output from the region for storing the predetermined charges. The imaging device according to item 18. 20. The charge signal correction means has a histogram detection circuit for detecting a variation amount of a signal level distribution of the plurality of output signals output from the region for storing the predetermined charge. 18 or claim 1
9. The imaging device according to item 9. 21. The charge signal correction means has an averaging correction circuit based on a variation amount detected from the plurality of output signals output from the region for storing the predetermined charges. To any one of claims 20 to 20
The imaging device according to the item. 22. The charge signal correction means has a gain correction circuit based on the amount of fluctuation detected from the plurality of output signals output from the region for storing the predetermined charges. Claim 1
The imaging device according to the item. 23. The image pickup apparatus according to claim 1, further comprising an image synthesizing unit that synthesizes the plurality of output signals corrected by the division signal correcting unit.