JP2007295260A - Image processing method and digital camera - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing method and a digital camera which calculate appropriate black level from a luminance histogram of an optical black area, even when the digital camera has high sensitivity with respect to increase in the noise. <P>SOLUTION: The luminance values outputted from pixels in the optical black area of an imaging face of a solid-state imaging element are used to generate a luminance histogram, wherein negative luminance values are clipped to zero luminance (step S100). When the ISO sensitivity of the digital camera is set to a high sensitivity, equal to or higher than a prescribed value in a step S102, the luminance values are sorted in the descending order (step S106), and a center value is extracted (step S108), and the center value is adopted as the black level. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an image processing method for performing black level correction and a digital camera.

  The imaging surface of a solid-state imaging device in which pixels (light receiving devices) are arranged in a two-dimensional form, such as a charge coupled device (CCD) type and a metal oxide semiconductor (MOS) type, is usually It is divided into an effective pixel area that receives light from the object scene and an optically shielded area around that area, that is, an optical black (OB) area. The optical black region is a portion where pixels exist but light is blocked due to the structure of the solid-state imaging device.

  The optical black region is often provided in a digital camera in order to grasp the luminance of a pixel output in the absence of light. Therefore, the average value of the luminance of the optical black region is generally called “black level” in the sense of “the luminance of the pixel without light”.

  This black level fluctuates each time an image is taken for reasons such as temperature characteristics. Generally, when the temperature is high, the luminance of the pixels on the entire imaging surface is increased, so that the black level is high, and when the temperature is low, the black level is low. In recent digital cameras, in order to absorb such black level fluctuations, as a first step, the entire screen including the optical black area is imaged with a certain offset in advance, and as a second step, optical black is captured. In many cases, the average value of the area is calculated to obtain the black level at the time of shooting, and is subtracted from the signal of the entire screen. Such a process is called black level correction in the text.

According to Patent Document 1 and Patent Document 2, by correcting a defective pixel (such as a defective pixel) existing in the optical black region, the black level calculated as the luminance average of the optical black region is further increased. A technique for accurately obtaining and improving the accuracy of subsequent image processing such as black level correction is disclosed.
JP 2001-145026 A JP 2004-350104 A

  In a digital camera, in high-sensitivity shooting such as setting ISO (International Organization for Standardization) sensitivity high, luminance noise output from a pixel increases. The same applies to the noise generated in the optical black region, and the number of pixels with negative luminance increases. However, since the brightness having negative luminance cannot be expressed when determining the black level, the luminance of the pixel in the optical black area having a negative value is clipped to luminance zero. That is, as the sensitivity of the digital camera is set higher, the number of pixels clipped to zero brightness increases. If the average value of the luminance histogram clipped to zero luminance is taken and set as the black level, a value higher than the original average value is calculated as the average value, and this is set as the black level. As a result, the subsequent image processing including black level correction may be affected.

  Patent Document 1 and Patent Document 2 mention correction of defective pixels existing in the optical black region, but as described above, the number of pixels clipped to zero brightness during high-sensitivity shooting increases, and as a result. The black level cannot be calculated accurately, and as a result, no particular mention is made of the adverse effect on the subsequent image processing.

  In view of such problems, the present invention provides an image processing method and a digital camera capable of accurately calculating a black level even when the ISO sensitivity is high and appropriately performing subsequent image processing such as black level correction. With the goal.

  In order to solve the above-described problems, an image processing method according to the present invention includes a step of creating a luminance histogram of pixels in an optical black region in an imaging surface of a solid-state imaging device, and a negative luminance among the created luminance histograms. A step of clipping to zero, a step of setting a median value of the clipped luminance histogram to a black level, and a step of correcting a black level of an image generated from a pixel on the basis of the black level.

  Alternatively, instead of the black level described above, the average value of the remaining histograms obtained by removing the frequency of zero luminance and the frequency corresponding to the frequency of zero luminance in descending order from the clipped luminance histogram may be used as the black level. Good.

  Alternatively, instead of the above-described black level, an average value of only the positive luminance histograms having a frequency equal to or greater than a predetermined number may be used as the black level. At this time, the predetermined number may be the maximum frequency, and thereby the positive luminance having the maximum frequency may be the black level.

  Alternatively, instead of the above-described black level, a low-pass filter may be applied to the clipped luminance histogram, and the positive luminance having the highest frequency in the histogram subjected to the low-pass filter may be set as the black level.

  According to the present invention, even when the sensitivity of a digital camera is high and noise increases, it is expected that black level can be accurately calculated and subsequent image processing such as black level correction can be appropriately executed. it can.

  Next, embodiments of an image processing method and a digital camera according to the present invention will be described in detail with reference to the accompanying drawings. FIG. 1 is a block diagram of a digital camera 10 which is an embodiment of the present invention. The digital camera includes an imaging unit 12, which may be a solid-state imaging device such as a charge coupled device (CCD) type or a complementary metal oxide semiconductor (CMOS) type. FIG. 2 is a schematic diagram of the imaging surface of the imaging unit 12 shown in FIG. As shown in FIG. 2, pixels (light receiving elements) are two-dimensionally arranged on the imaging surface of the imaging unit 12. The pixel arrangement method may be a lattice arrangement method or a honeycomb arrangement method. The arrangement of the color filters may be any other method such as the Bayer method. The imaging unit 12 shown in FIG. 2 has an effective pixel region 14 located above the imaging surface and an optical black region 16 located below. In this embodiment, the effective pixel area and the optical black area have a vertical positional relationship, but the optical black area may be located on either side of the effective pixel area, and the effective pixel area is oriented from four directions. You may do it.

  The effective pixel area 14 is an area in which pixels that perform photoelectric conversion of light from the object scene and image a subject are arranged. On the other hand, the optical black region 16 is a region that is optically shielded, and is a region where pixels are present but photoelectric conversion is not performed. The average luminance value of the optical black area 16 is generally the black level.

  FIG. 3 is a luminance histogram output from the optical black area shown in FIG. FIG. 3A is a histogram when the sensitivity of the digital camera is relatively low, and FIG. 3B is a histogram when the sensitivity of the digital camera is relatively high. As shown in FIG. 3A, when the camera sensitivity is low, the noise is small, so that the distribution variance is small, and the luminance histogram is normally a normal distribution with a positive value. Therefore, if the average value a of the normal distribution is calculated, this may be adopted as the black level.

  On the other hand, when the sensitivity of the camera is high, noise increases in the pixel group in the optical black region, and the variance of the luminance distribution increases as shown in FIG. Accordingly, a part of the distribution has a negative value as indicated by the dotted line portion A, and the negative luminance is clipped to the luminance zero as indicated by the symbol B. In this case, the black level should be the value b, which is the average value before clipping, and the average value of the histogram after clipping is the value c. That is, the value c that is higher than the value b that should be adopted as the black level is calculated as an average value and is adopted as the black level. If black level correction is performed in which the floating average value c is regarded as the black level and subtracted from the signal of the entire screen, an excessive value is subtracted as the black level, which causes problems such as darkening of the entire image. there is a possibility.

  Returning to FIG. 1, the description of the configuration of the digital camera 10 will be continued. The camera 10 includes a signal processing unit 20, which is a device that processes the luminance signal arriving from the above-described imaging unit 12 to appropriately determine the black level and performs various image processing such as black level correction. . The description of other elements included in the digital camera 10 will be described later. First, the processing contents of the signal processing unit 12 will be described below.

  FIG. 4 is a histogram showing a processing example of the signal processing unit shown in FIG. FIG. 4A shows a histogram before negative luminance is clipped to zero brightness, and FIG. 4B shows a histogram after negative luminance is clipped to zero luminance. For simplicity, the number of pixels is set to 20, but in reality, there are hundreds of thousands to more than 10 million pixels.

  The characteristic of the black level calculation of the signal processing unit 20 in the present embodiment is that the median (median) of the luminance histogram is set to the black level. According to such a method, even when the camera 10 is set to high sensitivity, noise increases, the variance of the luminance histogram increases, and the number of pixels clipped to the luminance zero increases, the average value is simply used as the black level. It is expected that an appropriate value can be adopted as the black level. In the present invention, it is assumed that the luminance distribution is a substantially symmetrical distribution such as a normal distribution.

  The luminance of the pixels in the optical black area before the negative luminance is clipped to zero is -3, -2, -1, -1, 0, 0, 0, 1 as shown in FIG. , 1, 1, 1, 1, 1, 2, 2, 2, 3, 3, 4, 5 (20 in total). The average value of the luminance at this point is calculated as an addition average, and 20/20 = 1. On the other hand, the luminance of the pixels in the optical black area after the negative luminance is clipped to zero is 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 2. , 2, 2, 3, 3, 4, 5. If the average value of the 20 luminances after the clipping is the black level, 27/20 = 1.35> 1, and a value exceeding 1 which is the original average value is output as the average value. However, if the median value of median (Median) is taken, a brightness value of 1 is obtained, which matches the original average value before clipping.

  FIG. 5 is a histogram showing another processing example of the signal processing unit shown in FIG. FIG. 5A shows a histogram before negative luminance is clipped to zero luminance, and FIG. 5B shows a histogram after negative luminance is clipped to zero luminance. For simplicity, the number of pixels is also 20 in this embodiment. It is assumed that the luminance distribution in the optical black region before and after the negative luminance is clipped to zero is the same as that in FIG. 4 described in the first embodiment.

  The characteristic of the black level calculation of the signal processing unit 20 in the present embodiment is that the average of the remaining histograms obtained by removing the frequency of zero luminance and the frequency corresponding to the frequency of zero luminance in descending order from the clipped luminance histogram. The value is a black level. In the first embodiment, the median value is adopted as the black level. However, in this method, the median value is not necessarily the average value of the distribution before clipping, for example, the median value becomes the maximum value or the minimum value. An inappropriate value may be adopted as the level. On the other hand, according to the present embodiment, since the average value of a plurality of luminances that are distributed at the center of the luminance distribution is set to the black level, the average value of the distribution after clipping is simply used as the black level, It can be expected that a more appropriate value can be adopted as the black level even when compared with the first embodiment employing the value.

  As shown in FIG. 5B, the frequency of the pixel clipped to zero brightness is 7 as indicated by the symbol D. On the other hand, the luminances of these 20 pixels are sorted and arranged in the order of high luminance, and the luminances of seven pixels having the same number as the frequency of zero luminance are listed in order from the pixel having the highest luminance, which is denoted by symbol E As shown, 5, 4, 3, 3, 2, 2, 2. Therefore, if the frequencies of the seven pixels having zero luminance and the frequencies of the seven pixels in order of increasing luminance are excluded from the distribution, the luminance distribution of the remaining pixels is 1, 1, as shown in FIG. 1, 1, 1, 1. These average values are 1, which coincides with 1, which is the average value of the luminance distribution before clipping.

  As described above, in this embodiment, the frequency X of the luminance zero pixel in the luminance distribution after clipping and the frequency X from the pixel having the maximum luminance to the X-th pixel in order are added to a total of 2X pixels. Are excluded from the luminance distribution of the pixels in the optical black region, and the average luminance of the remaining pixels is set as the black level. As a result, it is expected that an appropriate black level with no average value will be obtained.

  FIG. 6 is a histogram showing another processing example of the signal processing unit shown in FIG. FIG. 6A shows a histogram before negative luminance is clipped to zero luminance, and FIG. 6B shows a histogram after negative luminance is clipped to zero luminance. For simplicity, the number of pixels is also 20 in this embodiment.

  A characteristic of the black level calculation of the signal processing unit 20 in the present embodiment is that an average value of only positive luminance histograms having a frequency equal to or higher than a predetermined number among the clipped luminance histograms is set as the black level.

  Assuming that the luminance histogram of the optical black region before the negative luminance is clipped to the luminance zero is as shown in FIG. 6 (a), the luminance 2 is the average value of the histogram before clipping. It can be said that it is appropriate to regard it as a level. Therefore, in FIG. 6B showing the luminance histogram of the optical black region after clipping, in this embodiment, the average of pixels having a frequency other than zero luminance and having a frequency equal to or higher than a predetermined threshold value 4 indicated by a symbol F is used. The value is the black level.

  According to FIG. 6 (a), the luminance of the pixel before clipping is -2, -1, -1, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2, 2, 3, 3, 3, 3, 4, 4, 4, 5, 5, 6. The average value of this distribution is 2, and as shown in FIG. 6 (b), when negative luminance is clipped to luminance zero, 0, 0, 0, 0, 0, 0, 1, 1, 1, The distribution is 1, 2, 2, 2, 2, 2, 3, 3, 3, 3, 4, 4, 4, 5, 5, 6. Among these, the luminance of a pixel having a positive luminance with a frequency of 4 or more is three types of luminances 1, 2 and 3. If the average value of the frequencies of these luminances 1, 2, and 3 is calculated, it is 2, which matches the average value before clipping.

  Note that the threshold frequency may be set freely. The threshold value may be the maximum frequency. In that case, the luminance of the pixel having the maximum frequency is directly at the black level. In the example shown in FIG. 6B, this corresponds to adopting the luminance 2 having the maximum frequency 5 as the black level as it is.

  FIG. 7 is a histogram showing another variation of the present embodiment. In this embodiment, when the maximum frequency is set as a threshold and the luminance having the frequency is set as the black level as it is, the maximum value may be set as the black level. This corresponds to adopting the maximum value indicated by the symbol H as the black level, where the luminance indicated by the symbol G in FIG. Therefore, as shown in FIG. 7B, a high-frequency component is removed by applying a low-pass filter to the luminance histogram to create a smooth distribution as indicated by reference numeral I. As a result, it is possible to avoid the local maximum being erroneously adopted as the black level as described above.

  If a low-pass filter is applied to the luminance histogram as in this embodiment, even if the median value is set to the black level as in the first embodiment, the value falling into the maximum value is mistakenly set to the black level. May be avoided.

  Next, returning to FIG. 1, other elements of the digital camera for executing the first to third embodiments will be briefly described. The digital camera 10 has an optical system 30, which is a part that captures light from the object scene. Although not shown, the optical system 30 includes a lens, a diaphragm, and a shutter, and has an AE / AF (Auto Exposure / Auto Focus) drive function. AE / AF drive function controls exposure and adjusts focus.

  An image signal that is taken in from the optical system 30 into the imaging unit 12 and generated by photoelectric conversion is output to a preprocessing unit 32 connected to the imaging unit 12. Although not shown in detail, the pre-processing unit 32 performs correlated double sampling (CDS), automatic gain control (AGC) analog-to-digital conversion (Analog-to-analog) on the image signal. -Digital Conversion) and other preprocessing devices.

  The digital camera 10 includes a buffer memory 34, which is a recording device that is connected to the preprocessing unit 32 and temporarily stores the image signal digitized by the preprocessing unit 32. A volatile recording device may be used, and a recording medium such as an optical recording device or a magnetic recording device may be used. The buffer memory 34 can output the recorded digital image signal to the data bus 58.

  In addition to the function of determining the black level already described with respect to the first to third embodiments, the signal processing unit 20 floats and captures the signal of the entire screen including the optical black area by a certain offset. Further, the apparatus performs black level correction, white balance correction, gamma correction, and the like for subtracting the determined black level from the luminance signal of the entire screen.

  FIG. 8 is a functional block diagram showing an example of processing performed in the signal processing unit 20. The signal processing unit 20 first performs the above black level correction 40 on the input image signal. This is because the purpose is to appropriately select a black level in black level offset correction and perform black level correction. However, the black level correction may be performed in a later process. The white balance correction 42 is a color correction process for reproducing an appropriate white color following the change in the color temperature of the light source and maintaining normal color reproducibility. The linear matrix process 44 is a process for converting R (Red), G (Green), and B (Blue) signals into luminance / color difference signals using a linear matrix. The gamma correction 46 is a correction operation for adjusting the relative relationship between the image color data and the signal when it is actually output to obtain a more natural display. The synchronization process 48 is a process of interpolating the three primary colors of RGB to each pixel to which any color filter is assigned. The contour correction 50 is a correction process for clearly reproducing the contour of the subject.

  Returning again to FIG. 1, the description of the other elements of the camera 10 will be continued. The monitor 52 is a display device such as a liquid crystal display (LCD), and outputs an image subjected to various processing by the signal processing unit 20. The camera 10 also includes a storage 54, which is a recording device that records the digital image data processed by the signal processing unit 20. The storage 54 is a storage device connected to the data bus 58 via the storage interface circuit 56, and is a device that finally stores the image signal that has been subjected to image processing by the signal processing unit 20. The system control unit 60 is a processor that controls the entire camera 10, and controls the timing signal generator 62 connected thereto to generate a timing pulse, which is given to a driver for driving the imaging unit, and the imaging unit Drive. Further, the system control unit 60 gives a control signal to the preprocessing unit 32 directly or indirectly through the timing signal generator 62 to control various processes of the preprocessing unit 32.

  The camera 10 further includes an operation unit 66, which has a shutter release button for imaging and other operation buttons, and is a part for operating the camera 10 by giving an operation signal to the system control unit 60. In particular, in the present embodiment, the operation unit 66 has a function of setting the ISO sensitivity of the camera 10, and the system control unit 60 causes the imaging unit 12 to perform imaging according to the set ISO sensitivity. The set ISO sensitivity is also given to the signal processing unit 20 through the data bus 58, and affects the black level determination by the signal processing unit 20.

  The operation of the digital camera according to the embodiment of the present invention configured as described above will be described below. FIG. 9 is a flowchart showing the operation of the first embodiment of the present invention. The ISO sensitivity is transmitted in advance to the system control unit 60 by the operation of the operation unit 66 shown in FIG. When imaging is executed by pressing the shutter release button of the operation unit 66, the imaging unit 12 given light from the object scene from the optical system 30 generates an image signal by photoelectric conversion, The data is sent to the processing unit 32. The image signal that has been subjected to various processes in the preprocessing unit 32 and digitized is output to the data bus 58 via the buffer memory 34 and input to the signal processing unit 20. When the first embodiment is adopted, the signal processing unit 20 first creates a luminance histogram of the optical black region 16 in the input digital image signal as shown in step S100. When the histogram is as shown in FIG. 4A, for example, at the time of step S100, the negative luminance is clipped to zero luminance and changed to the histogram shown in FIG. 4B.

  Next, the signal processing unit 20 knows the preset ISO sensitivity from the system control unit 60, and if the ISO sensitivity is set to a low sensitivity smaller than a predetermined value as shown in step S102, the process goes to step S104. Then, the average value of the clipped luminance histogram is calculated and adopted as the black level, and black level correction and other subsequent processing shown in FIG. 8 are performed. In this case, it is considered that since the sensitivity is low, there is little noise, the variance of the luminance histogram is small, and the negative luminance is often clipped to zero luminance.

  On the other hand, if the ISO sensitivity is set to a high sensitivity equal to or higher than a predetermined value in step S102, the process proceeds to step S106, and the luminances are sorted in descending order. Next, in step S108, the median is extracted and adopted as the black level. In this case, since the digital camera 10 is set to high sensitivity, a lot of noise is generated and it is considered that there are many pixels clipped to zero brightness.

  FIG. 10 is a flowchart showing the operation of the second embodiment of the present invention. Steps S200, S202, and S204 are the same as steps S100, S102, and S104 in FIG. If the ISO sensitivity is equal to or higher than the predetermined value, in step S206, X frequencies with zero luminance are excluded from the luminance histogram. That is, when the luminance histogram in which the negative luminance is clipped to zero luminance is as shown in FIG. 5B, the frequency indicated by the symbol D in FIG. 5 is excluded from the histogram. Then, in the next step S208, the luminances are sorted in descending order, and in the next step S210, X frequencies having the same number as the luminance of zero luminance are excluded in the sorted large luminance order. That is, the frequency indicated by the symbol E is excluded from the histogram of FIG. In step S204, the luminance average value of the remaining histogram, that is, the histogram shown in FIG. 5C is calculated and adopted as the black level.

  FIG. 11 is a flowchart showing the operation of the third embodiment of the present invention. Steps S300, S302, and S304 are the same as steps S100, S102, and S104 in FIG. If the ISO sensitivity is equal to or higher than the predetermined value, in step S306, frequencies other than the luminance of the predetermined frequency or higher or the maximum frequency are excluded from the histogram. This is because, for example, when a luminance histogram in which negative luminance is clipped to zero luminance is as shown in FIG. 6B, a portion other than the portion indicated by the symbol F having a frequency of a predetermined frequency of 4 or more is histogrammed. It means to exclude from. Alternatively, frequencies other than the luminance having the maximum frequency 5 are excluded from the histogram. In the histogram from which the frequency is excluded in this way, an average value of luminance is calculated as shown in step S304, and this is adopted as the black level. In step S306, when the luminance other than the maximum frequency is excluded, the positive luminance 2 having the maximum frequency 5 becomes the black level as it is.

1 is a block diagram of a digital camera that is an embodiment of the present invention. FIG. It is a schematic diagram of the imaging surface of the imaging part shown in FIG. FIG. 2 is a histogram of luminance output from the optical black area shown in FIG. 2, where (a) is a histogram when the sensitivity of the digital camera is relatively low, and (b) is a histogram when the sensitivity of the digital camera is relatively high. . It is a processing example of the signal processing part shown in FIG. 1, and is a histogram showing the first embodiment of the present invention. It is a processing example of the signal processing part shown in FIG. 1, Comprising: It is a histogram which shows the 2nd Example of this invention. It is a processing example of the signal processing part shown in FIG. 1, Comprising: It is a histogram which shows the 3rd Example of this invention. It is a histogram which shows the other variation of the 3rd Example demonstrated in FIG. It is a functional block diagram which shows an example of the process performed in the signal processing part shown in FIG. It is a flowchart which shows operation | movement of the 1st Example shown in FIG. It is a flowchart which shows operation | movement of the 2nd Example shown in FIG. It is a flowchart which shows operation | movement of the 3rd Example shown in FIG.

Explanation of symbols

10 Digital camera
12 Imaging unit
14 Effective pixel area
16 Optical black area
20 Signal processor
60 System controller
66 Control panel

Claims (10)

Creating a luminance histogram of pixels in the optical black region of the imaging surface of the solid-state imaging device;
Clipping negative luminance of the created luminance histogram to zero luminance;
Setting the median value of the clipped luminance histogram to a black level;
And a step of correcting a black level of an image generated from the pixel with reference to the black level. Creating a luminance histogram of pixels in the optical black region of the imaging surface of the solid-state imaging device;
Clipping negative luminance of the created luminance histogram to zero luminance;
Removing the frequency of zero brightness from the clipped luminance histogram and the frequency corresponding to the frequency of zero brightness in descending order of brightness, and setting the average value of the remaining histogram to a black level;
And a step of correcting a black level of an image generated from the pixel with reference to the black level. Creating a luminance histogram of pixels in the optical black region of the imaging surface of the solid-state imaging device;
Clipping negative luminance of the created luminance histogram to zero luminance;
A step of setting an average value of only positive luminance histograms having a frequency equal to or higher than a predetermined number among the clipped luminance histograms to a black level;
And a step of correcting a black level of an image generated from the pixel with reference to the black level.   4. The image processing method according to claim 3, wherein the predetermined number is a maximum frequency, and thereby a positive luminance having the maximum frequency is set to a black level. Creating a luminance histogram of pixels in the optical black region of the imaging surface of the solid-state imaging device;
Clipping negative luminance of the created luminance histogram to zero luminance;
Applying a low pass filter to the clipped luminance histogram;
Of the histogram subjected to the low-pass filter, a step of setting a positive luminance having the maximum frequency to a black level;
And a step of correcting a black level of an image generated from the pixel with reference to the black level. A solid-state imaging device;
Means for creating a luminance histogram of pixels in the optical black region of the imaging surface of the solid-state imaging device;
Means for clipping the negative luminance of the created luminance histogram to zero luminance;
Means for setting the median value of the clipped luminance histogram to a black level;
Means for correcting a black level of an image generated from the pixel with reference to the black level. A solid-state imaging device;
Means for creating a luminance histogram of pixels in the optical black region of the imaging surface of the solid-state imaging device;
Means for clipping the negative luminance of the created luminance histogram to zero luminance;
Means for removing the frequency of zero brightness from the clipped luminance histogram and the frequency corresponding to the frequency of zero brightness in descending order of brightness, and setting the average value of the remaining histogram as a black level;
Means for correcting a black level of an image generated from the pixel with reference to the black level. A solid-state imaging device;
Means for creating a luminance histogram of pixels in the optical black region of the imaging surface of the solid-state imaging device;
Means for clipping the negative luminance of the created luminance histogram to zero luminance;
Means for setting a black level to an average value of only positive luminance histograms having a frequency equal to or greater than a predetermined number among the clipped luminance histograms;
Means for correcting a black level of an image generated from the pixel with reference to the black level.   9. The digital camera according to claim 8, wherein the predetermined number is a maximum frequency, whereby a positive luminance having the maximum frequency becomes a black level. A solid-state imaging device;
Means for creating a luminance histogram of pixels in the optical black region of the imaging surface of the solid-state imaging device;
Means for clipping the negative luminance of the created luminance histogram to zero luminance;
Means for applying a low pass filter to the clipped luminance histogram;
Means for setting a positive luminance having the maximum frequency in the histogram subjected to the low-pass filter to a black level;
Means for correcting a black level of an image generated from the pixel with reference to the black level.