JP2010193099A - Image capturing apparatus and method of controlling the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image capturing apparatus which can perform image capture with a dynamic range in consideration of a subject and a scene. <P>SOLUTION: The image capturing apparatus determines a scene of a captured image, and, depending on the scene decision results, performs image capturing with an expanded dynamic range, or performs a dynamic range reduction processing based on the captured image. In executing the dynamic range expansion processing, the image capturing apparatus performs image capturing at a decreased imaging sensitivity and performs tone correction for compensating the decreased imaging sensitivity with respect to the captured image. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an imaging apparatus and a control method thereof.

  Conventionally, digital cameras and digital videos using image sensors such as CCD image sensors and CMOS image sensors have been widely used. However, these image sensors have a narrow dynamic range (latitude) compared to silver salt films. For this reason, when a high-contrast scene is imaged, it is easy to cause gradation deterioration (blackout) in a low luminance part and gradation deterioration (whiteout) in a high luminance part.

  In response to such a problem, a system capable of automatically controlling the dynamic range has been proposed.

  For example, in Patent Document 1, when a main subject is detected to be backlit or high-contrast from a captured image, a black saturation point and a white saturation point are specified from the histogram of the image, and the brightness of the main subject is determined. It has been proposed to perform gradation correction so as to be appropriate.

  Moreover, in patent document 2, using the imaging device which has arrange | positioned the high sensitivity light receiving element and the low sensitivity light receiving element, the same scene is imaged with the light receiving element of different sensitivity, and two types of images with different dynamic ranges are acquired, It has been proposed to synthesize them according to the scene analysis results.

  There is also a digital camera that has an imaging mode (high brightness side / gradation priority mode) that suppresses whiteout in high brightness areas by shifting the sensitivity setting range by one step to the high sensitivity side (Canon, EOS 5D). Mark2).

Japanese Patent Application Laid-Open No. 2005-209012 JP 2004-186876 A

  However, in the method of Patent Document 1, although there is an effect for enhancing the contrast feeling, the point at which the sensor is saturated does not change, so the effect of expanding the dynamic range cannot be obtained.

  Further, although the method of Patent Document 2 can obtain the effect of expanding the dynamic range, there is a problem that it is expensive because it is necessary to use a special imaging device in which light receiving elements having different sensitivities are arranged.

  Further, in a digital camera having a gradation priority mode, in order to obtain an imaging result in which whiteout in a high-luminance portion is suppressed, it is necessary for the user to explicitly set the gradation priority mode for imaging. Depending on the scene, it may be more difficult to feel the effect of the gradation priority mode, or it may be better to take an image with priority given to other conditions than suppression of whiteout in a high luminance part. For example, when the subject is a landscape, it is desirable that there is no whiteout in the high luminance part. However, when the subject is a person, the brightness of the person's face is appropriate even if the background is slightly overexposed. It is more desirable. That is, in order to obtain an appropriate imaging result by using the gradation priority mode, it is required that the user can appropriately determine the scene in addition to setting the gradation priority mode.

  The present invention has been made in view of the problems of the prior art. An object of the present invention is to provide an imaging apparatus capable of imaging in a dynamic range in consideration of a subject or a scene and a control method thereof.

  The above-described object is a scene determination unit that performs scene determination based on a moving image that is captured at the time of standby for still image capturing, and a control unit that controls the image capturing operation. A dynamic range expansion process that applies gradation correction to compensate for a reduction in imaging sensitivity for images taken with still images, and still images, moving images or still images And a control unit that executes any one of a dynamic range reduction process that applies gradation correction based on the captured image to a still image captured image.

  With such a configuration, according to the present invention, it is possible to realize an imaging apparatus capable of imaging in a dynamic range in consideration of a subject or a scene and a control method thereof.

It is a block diagram which shows the structural example of the imaging device which concerns on embodiment of this invention. It is the figure which showed typically the process in which a face is detected from an original image. It is a flowchart for demonstrating the face detection operation | movement in the face detection circuit of the imaging device which concerns on embodiment of this invention. It is a chromaticity diagram showing representative colors in the CIE-L * a * b * color space. It is a figure which shows the example of the coefficient of the two-dimensional high pass filter which the face detection circuit of the imaging device which concerns on embodiment of this invention uses. It is a block diagram showing an example of composition of an AFE circuit of an imaging device concerning an embodiment of the present invention. It is a figure which shows typically the example of the histogram which the histogram preparation circuit of the imaging device which concerns on embodiment of this invention produces. It is the block diagram which showed the flow of the signal processing at the time of imaging standby in the imaging device which concerns on embodiment of this invention. It is a flowchart for demonstrating the determination operation | movement of the dynamic range (D range) expansion amount in the imaging device which concerns on embodiment of this invention. In the imaging device concerning an embodiment of the present invention, it is a figure showing the example of the D range expansion amount finally determined according to the size of the D range expansion amount of the face area, and the D range expansion amount of the whole image. It is a figure showing the conceptual diagram of D range in the imaging device concerning an embodiment of the present invention. It is a figure which shows the example of a relationship between AE target value, a saturation signal value, and D range in the imaging device which concerns on embodiment of this invention. It is a figure which shows the example of a setting of the gradation characteristic in the signal processing circuit of the imaging device which concerns on embodiment of this invention. It is a flowchart explaining operation | movement of this imaging process in the imaging device which concerns on embodiment of this invention. It is a figure which shows the example of an area | region division at the time of backlight determination in the imaging device which concerns on embodiment of this invention. In the imaging device which concerns on embodiment of this invention, the example of the gradation characteristic selected when a night view is determined is shown. It is a flowchart for demonstrating the determination operation | movement of the dynamic range reduction amount (D-) in the imaging device which concerns on embodiment of this invention. FIG. 18 is a diagram illustrating a relationship example between the final gain value Gain_yhp calculated in S504 of FIG. 17 and the final gain increase amount according to the imaging sensitivity in the imaging apparatus according to the embodiment of the present invention. FIG. 6 is a diagram illustrating a relationship between a gain increase amount and a gradation characteristic setting value in the imaging apparatus according to the embodiment of the present invention. Example of relationship between exposure control, dynamic range expansion processing (D +), dynamic range reduction processing (D−), and setting of gradation characteristics for night scene according to the scene determination result in the imaging apparatus according to the embodiment of the present invention. FIG.

Hereinafter, the present invention will be described in detail based on exemplary embodiments with reference to the accompanying drawings.
FIG. 1 is a block diagram illustrating a configuration example of an imaging apparatus according to an embodiment of the present invention. The imaging apparatus according to the present embodiment includes any apparatus having a function of capturing an image using an imaging element. Such devices include not only digital still cameras and digital video cameras, but also mobile phones, PDAs, personal computers, etc. that have built-in or connected cameras.

  In FIG. 1, the operation unit 112 includes buttons, switches, and the like, and is used by a user to give an instruction to the imaging apparatus or make settings. A shutter button is also included in the operation unit 112, and in this embodiment, it is possible to detect a half-pressed state and a fully-pressed state of the shutter button.

  The system controller 107 recognizes the half-pressed state of the shutter button as an imaging preparation start instruction and the fully-pressed state as an imaging start instruction. The system controller 107 includes, for example, a CPU, a ROM, a RAM, and the like, and controls the overall operation of the imaging apparatus when the CPU executes a program stored in the ROM using the RAM.

  The attitude sensor 133 detects the attitude of the imaging device and outputs the detection result to the system controller 107. The system controller 107 determines whether the posture of the imaging device is the vertical position or the horizontal position based on the output from the posture sensor 133. Note that the attitude sensor 133 may determine whether the position is the vertical position or the horizontal position, and the determination result may be output to the system controller 107.

  The lens device 200 includes a lens group including a focus lens, a drive device that drives the focus lens, a diaphragm, and a mechanical shutter, and operates based on control of the system controller 107.

  The imaging element 101 is a photoelectric conversion element such as a CCD image sensor or a CMOS image sensor. An analog front end (AFE) circuit 150 performs gain adjustment, A / D conversion, and the like on the analog image signal output from the image sensor 101 and outputs the digital image signal. Details of the AFE circuit 150 will be described later.

The buffer memory 103 temporarily stores the digital image signal output from the AFE circuit 150.
The compression / decompression circuit 104 encodes the captured image data into a recording image file (for example, JPEG file) format, or decodes the image file read from the recording medium 106.

The recording device 105 reads / writes data from / to a recording medium 106 such as a built-in memory or a removable memory card under the control of the system controller 107.
The display control circuit 108 controls the display operation on the display unit 110 having a display device such as an LCD in accordance with the control of the system controller 107.
The D / A converter 109 converts the display digital image signal output from the display control circuit 108 into an analog image signal that can be displayed by the display unit 110.

  The display unit 110 displays not only the captured image but also a GUI screen for the user to make various settings and instructions to the imaging device, and various information related to the imaging device. In addition, by sequentially displaying continuously captured images on the display unit 110, the display unit 110 can function as an electronic viewfinder (EVF). This sequential imaging / display operation for causing the display unit 110 to function as an EVF is also referred to as a through display or a live view display. This sequential imaging operation is substantially equivalent to the moving image imaging operation, and the actual image captured and displayed for EVF is a moving image.

  The face detection circuit 120 performs face detection as an example of a technique for detecting a person from a captured image. The face detection circuit 120 performs face detection processing on image data in YUV format or RAW format stored in the buffer memory 103, and generates a face detection result including the size and position of the face area in the image as a histogram generation circuit. To 130.

  The face detection method used by the face detection circuit 120 is not particularly limited, and any known method can be applied. As a known face detection technique, a method based on learning using a neural network or the like, template matching is used to search a part having a characteristic shape of eyes, nose, mouth, etc. from an image, and if the degree of similarity is high, it is regarded as a face There are methods. In addition, many other methods have been proposed, such as a method that detects image feature amounts such as skin color and eye shape and uses statistical analysis. A plurality of these methods can be combined to improve the accuracy of face detection. Specific examples include a face detection method using wavelet transform and image feature amount described in JP-A-2002-251380.

Here, a specific face detection operation example of the face detection circuit 120 will be described with reference to FIGS.
FIG. 2 is a diagram schematically showing the process of detecting a face from an original image, and FIG. 2A shows the original image.

FIG. 3 is a flowchart for explaining the face detection operation of the face detection circuit 120.
In S101, the face detection circuit 120 extracts a skin color area from the original image. FIG. 4 is a chromaticity diagram showing representative colors in the CIE-L * a * b * color space, and an ellipse therein is an area that is highly likely to be a skin color.
The face detection circuit 120 converts the original image data in RGB format into the L * a * b * format by a known method, and extracts a skin color region composed of pixels having chromaticity within the region indicated by the ellipse in FIG. . FIG. 2B schematically shows a skin color area extracted from the original image.

  Next, in S102, the face detection circuit 120 extracts a high frequency component from the extracted skin color region. Specifically, the face detection circuit 120 applies a high-pass filter to the skin color area. FIG. 5 is a diagram illustrating an example of coefficients of a two-dimensional high-pass filter. An example of an image obtained by applying a high-pass filter to the image of FIG. 2B is shown in FIG.

In S103, the face detection circuit 120 performs template matching using an eye template as shown in FIG. 2D on the image after the high-pass filter is applied, and detects an eye in the image.
In S104, the face detection circuit 120 determines the face area based on the positional relationship of the eye area detected in S103, and obtains a face detection result including the position and size of the face area.

  Returning to FIG. 1, the histogram creation circuit 130 acquires the detection result of the face area from the face detection circuit 120, and creates a histogram related to the luminance value of the pixels included in the face area. The histogram creation circuit 130 can also create a histogram of the luminance values of the included pixels for each partial region obtained by dividing the image into a plurality of regions. The created histogram is stored in the buffer memory 103.

  When acquiring color information for each partial area, the color evaluation value acquisition circuit 132 uses the partial area for the reduced image obtained by reducing the image analysis image for one screen stored in the buffer memory 103 by the resize circuit 131. Get color information for each. The color information may be color evaluation values such as saturation, hue, and luminance.

  The signal processing circuit 140 applies signal processing to the image data stored in the buffer memory 103 in accordance with signal processing parameters (such as white balance correction coefficients and gradation characteristic parameters) set by the system controller 107. Then, the signal processing circuit 140 generates YUV format image data and stores it again in the buffer memory 103.

  As will be described later, the imaging apparatus of the present embodiment realizes dynamic range control by sensitivity adjustment (gain adjustment) in the AFE circuit 150 and correction of gradation characteristics in the signal processing circuit 140.

FIG. 6 is a block diagram illustrating a configuration example of the AFE circuit 150.
The clamp circuit 151 clamps the signal output from the image sensor 101 to the reference black level so that the light shielding level of the sensor or the output value of the reference voltage region becomes zero.
The CDS gain circuit 152 applies a CDS gain (analog gain) to the clamped signal. The CDS gain applied in a general AFE circuit has discrete values such as 0, 3, 6 [dB].

  The signal to which the analog gain is applied is converted into digital data by the A / D converter 153. Next, a variable gain amplifier (VGA) gain circuit 154 applies a VGA gain to the digital data. In the present embodiment, it is assumed that the value of the VGA gain can be adjusted in units of 0.125 dB, for example, in the range of 0 to 36 dB. The signal to which the VGA gain is applied is clipped to a predetermined number of bits by the clip circuit 155 and output to the buffer memory 103.

  In this embodiment, the imaging sensitivity (ISO sensitivity) setting of the imaging apparatus is realized by controlling the values of the CDS gain and the VGA gain applied by the AFE circuit 150 by the system controller 107.

  Also, the sensitivity for correcting variations in the sensitivity characteristics of the image sensor 101 by irradiating the imaging device with a reference amount of light and controlling the CDS gain and VGA gain so that a constant luminance signal value is output from the imaging device. Adjustments can be made.

  In the present embodiment, sensitivity setting is performed by combining a CDS gain and a VGA gain. For example, when setting a gain of 6 dB as a whole at low sensitivity, for example, the CDS gain is set to 3 dB and the VGA gain is set to 3 dB. When a gain of 24 dB is set at high sensitivity, for example, the CDS gain is set to 6 dB and the VGA gain is set to 18 dB. As described above, since the CDS gain cannot generally be set finely, a rough gain is set by the preceding CDS gain, and the VGA gain is controlled in order to perform delicate sensitivity control such as a sensitivity adjustment portion.

  Generally, a gain circuit amplifies a noise component simultaneously with a signal component. Therefore, in order to suppress the amplification of the noise component superimposed in the analog circuit, it is preferable to set the preceding CDS gain as high as possible in the combination of the CDS gain and the VGA gain that can realize the total gain amount. Such a setting can also realize an effect that the quantization accuracy of the A / D converter 153 can be effectively utilized to the maximum extent.

Next, an operation at the time of imaging in the imaging apparatus having the above-described configuration will be described.
The image capturing apparatus according to the present embodiment captures a moving image and operates the display unit 110 as an EVF during standby in which the image capturing mode is operated and an image capturing preparation instruction or an image capturing start instruction is not input. That is, the system controller 107 performs processing for continuously capturing images at a predetermined rate (for example, 30 frames / second), generating a display image from the captured images, and displaying the display image on the display unit 110.

  When face detection execution is set, the face detection circuit 120 detects a face for a display image (hereinafter also referred to as an EVF image), and outputs the detection result to the system controller 107. Then, the system controller 107 instructs the display control circuit 108 together with the position information of the face area to superimpose and display the face frame for presenting the detected face area to the user on the EVF image.

  The face detection result is also supplied to the histogram creation circuit 130, which creates a histogram from pixels included in the face area in the EVF image. The histogram creation circuit 130 creates a histogram for each region obtained by dividing the entire image into a plurality of regions. The created histogram is stored in the buffer memory 103.

FIG. 7 is a diagram schematically showing an example of a histogram created by the histogram creation circuit 130 in the present embodiment.
FIG. 7 shows that a histogram 73 for each of the partial areas 71 obtained by dividing the entire image into four equal parts in the vertical and horizontal directions and a histogram 74 for the face area 72 are created. Note that the histograms 73 and 74 in FIG. 7 are cumulative histograms. Note that when creating a histogram for a partial region, the face region may be excluded. By doing so, it is possible to create a histogram of the face region and the rest (background).

  In the present embodiment, the luminance value YHi having a frequency of 80% in the cumulative histogram 73 of the partial region 71 and the luminance value YHiFace having a frequency of 90% in the cumulative histogram 74 of the face region 72 are as follows. It is used for evaluation of a whiteout area.

  When the user fully presses the shutter button and an imaging start instruction is input, the system controller 107 performs an imaging operation based on processing results such as automatic exposure control (AE) and automatic focus detection (AF). Specifically, the system controller 107 controls the focal position and aperture of the lens device 200, the mechanical shutter, the image sensor 101, and further, if necessary, a flash (not shown) to capture an image.

  The analog image signal output from the image sensor 101 is stored in the buffer memory 103 as digital image data through the AFE circuit 150 described above. The signal processing circuit 140 processes the digital image data in accordance with various signal processing parameters set by the system controller 107, generates image data in YUV format, and stores it again in the buffer memory 103.

  The image data processed by the signal processing circuit 140 is encoded into, for example, a JPEG format file by the compression / decompression circuit 104 and recorded on the recording medium 106 by the recording device 105.

  Further, the display control circuit 108 generates a display image from the YUV format image data stored in the buffer memory 103 and causes the display unit 110 to display it as a quick review image through the D / A converter 109.

  FIG. 8 is a block diagram illustrating the flow of signal processing during imaging standby in the imaging apparatus of the present embodiment. As described above, at the time of imaging standby, the imaging apparatus of the present embodiment performs continuous imaging and display for causing the display unit 110 to function as an EVF.

  The analog image signal output from the image sensor 101 is converted into gain data (sensitivity adjustment) and digital data by the AFE circuit 150. Then, so-called development processing such as pixel interpolation processing and white balance correction is performed on the RAW format image data by the signal processing circuit 140 to generate YUV format digital image data.

  The digital image data is stored in a display area (VRAM or display buffer) 103 a of the buffer memory 103, and is output to the display unit 110 through the display control circuit 108 and the D / A converter 109.

  On the other hand, the digital image data generated by the signal processing circuit 140 is also stored in the image analysis area (image analysis buffer) 103 b of the buffer memory 103. The image data stored in the image analysis buffer 103 b is used for face detection by the face detection circuit 120 and histogram creation by the histogram creation circuit 130. Note that it is not necessary to store all EVF image data in the image analysis buffer 103b, and a part of EVF image data corresponding to the period of face detection and histogram creation is stored.

  The face detection circuit 120 performs face detection on the image data stored in the image analysis buffer 103b. When a face is detected, information (for example, position and size) that can specify the face area is detected. The included face detection result is output.

  The histogram creation circuit 130 creates a histogram based on the image data stored in the image analysis buffer 103 b and the face detection result from the face detection circuit 120. As described above, the histogram creation circuit 130 creates the face area histogram 130a for the face area and the divided area histogram 130b for each area image obtained by dividing the entire image.

  The divided region histogram 130b can be created for each region image including the face region for the entire image, or can be obtained for each region image excluding the face region. The former is easy to process, but the latter is preferable in order to accurately detect whether a whiteout region is a face region or a background region.

  As described above, while the display unit 110 is functioning as an EVF, imaging is continuously repeated, and the VRAM is rewritten in a short time. In general, the time required for face detection and histogram creation is longer than the EVF image display period (for example, 1/30 second). Therefore, in this embodiment, an image analysis buffer 103b is provided in addition to the VRAM, and the image analysis buffer 103b is updated until face detection and histogram creation for the image data stored in the image analysis buffer 103b are completed. do not do.

  With this configuration, face detection and histogram creation can be performed on the same EVF image, and image analysis can be performed easily and with high accuracy. Of course, it may be performed if face detection and histogram creation are possible for each frame of the EVF image. However, since it is generally unlikely that the captured scene will change significantly from frame to frame, It is not necessary to execute. Therefore, the load on the system controller 107 can be reduced.

The image data stored in the image analysis buffer 103 b is reduced by the resizing circuit 131, and the color evaluation value acquisition circuit 132 acquires the color evaluation value.
For example, it is assumed that the image data stored in the image analysis buffer 103b corresponds to a VGA size image. The resizing circuit 131 first generates, for example, YUV 4: 2: 2 image data of horizontal 64 pixels and vertical 48 pixels from YUV 4: 2; 2 image data of horizontal 640 pixels and vertical 480 pixels. After that, the resizing circuit 131 averages Y in a block composed of 4 pixels of 2 pixels in the horizontal direction and 2 pixels in the vertical direction, averages the UV in 2 pixels in the vertical direction, and has a YUV 4 of 32 pixels in the horizontal direction and 24 pixels in the vertical direction. : 4: 4 Image data is generated.
Note that as the resizing method in the resizing circuit 131, other methods such as averaging in predetermined plural pixel units, simple resampling, thinning out by linear interpolation, bicubic interpolation, and the like may be used.

For the reduced image data in the horizontal 32 pixels, vertical 24 pixels, and YUV 4: 4: 4 format generated by the resizing circuit 131, the color evaluation value acquisition circuit 132 performs luminance information Y and hue information as color evaluation values for each block. H and saturation information C are calculated.
The color evaluation value acquisition circuit 132 can calculate a color evaluation value by the following formula, for example.
Y = Y
C = √ (U ² + V ² )
H = tan ⁻¹ (U / V)

Of course, the color evaluation value may be an evaluation value in the color space of CIELab or an evaluation value in another color space. The color evaluation value may also be calculated using a mathematical library that performs the above-described calculation, or may be calculated in a pseudo manner by referring to a lookup table prepared in advance.
In this way, luminance information, hue information, and saturation information can be acquired as color evaluation values for each block obtained by dividing the entire screen into 32 and 24.

FIG. 9 is a flowchart for explaining the operation of determining the dynamic range expansion amount (D +) in the imaging apparatus of the present embodiment.
In the present embodiment, the amount of whiteout in the captured image is obtained based on the face detection result and the histogram creation result for the EVF image, and the dynamic range expansion amount is determined according to the amount of whiteout. Then, the exposure and sensitivity at the time of actual imaging are adjusted using a dynamic range expansion amount determined in advance using the EVF image.

  In step S201, the system controller 107 continuously captures images as described above, generates an EVF image, and sequentially stores the EVF image from the signal processing circuit 140 in the display buffer 103a. The EVF image is also stored in the image analysis area (image analysis buffer) 103b of the buffer memory 103 at a predetermined cycle.

In S202, the face detection circuit 120 performs face detection on the EVF image stored in the image analysis buffer 103b.
When the face detection is successful (S203, Y), the histogram creation circuit 130 creates a face area histogram from the face area of the EVF image based on the face detection result from the face detection circuit 120 (S204).
In step S <b> 205, the system controller 107 serving as a whiteout amount calculation unit calculates a whiteout amount of the face area from the histogram of the face area.

  When face detection fails (S203, N), and after calculating the amount of whiteout in the face area, in S206, the histogram creation circuit 130 creates a divided area histogram for each area image obtained by dividing the entire EVF image. Here, as an example, a divided region histogram is created for each of 16 divided regions obtained by dividing the entire EVF image into four equal parts.

In step S207, the system controller 107 calculates the amount of whiteout in the entire EVF image from the divided region histogram.
In step S208, the system controller 107 serving as a dynamic range expansion amount determination unit determines a dynamic range expansion amount based on at least the amount of whiteout in the entire image obtained in step S207.

Next, a specific example of the whiteout amount calculation process performed by the system controller 107 in S205 and S207 of FIG. 9 will be described.
First, the calculation process of the amount of whiteout in the face area will be described.
In step S205, the system controller 107 calculates a luminance value YHiFace from which the cumulative frequency of the cumulative histogram becomes a predetermined value (90% in the example of FIG. 7) as the amount of whiteout in the face region from the face region histogram created in S204. To do.

  In step S208, the system controller 107 determines the dynamic range expansion amount (D + (face)) of the face area according to the magnitude relationship between the value of the whiteout amount YHiFace of the face area and a predetermined threshold value. To do.

Specifically, for example, when the predetermined threshold is set in three stages of THHiFace, THMidFace, and THLowFace in descending order,
D + (face) = 1 correction level (when YHiFace> THHiFace)
D + (face) = correction level 2/3 stage (when THHiFace ≧ YHiFace> THMidFace)
D + (face) = correction level 1/3 step (when THMidFace ≧ YHiFace> THLowFace)
D + (face) = correction level 0 stage (when THLowFace ≧ YHiFace)
And decide.

  In S206, the system controller 107 determines the luminance value at which the cumulative frequency of the cumulative histogram is a predetermined value (80% in the example of FIG. 7) as the amount of whiteout in the partial region for each of the divided region histograms created in S206. YHi_n (n is 1 to the number of divisions, 16 in the example of FIG. 7) is calculated.

  In S207, the system controller 107 counts the number YH_BNum of regions where the luminance value YHi_n exceeds a predetermined threshold Y_B_Th. Then, the system controller 107 determines the dynamic range expansion amount (D + (background)) of the entire image in accordance with the size relationship between the number of regions YH_BNum and a predetermined threshold value.

Specifically, for example, if the predetermined threshold value is ThYH_BNum6 to ThYH_BNum0 in order from the largest value,

D + (background) = correction level 6/3 stage (if YH_BNum> ThYH_BNum6)
D + (background) = correction level 5/3 stage (when ThYH_BNum6 ≧ YH_BNum> ThYH_BNum5)
D + (background) = correction level 4/3 stage (when ThYH_BNum5 ≧ YH_BNum> ThYH_BNum4)
D + (background) = correction level 3/3 stage (when ThYH_BNum4 ≧ YH_BNum> ThYH_BNum3)
D + (background) = correction level 2/3 stage (when ThYH_BNum3 ≧ YH_BNum> ThYH_BNum2)
D + (background) = correction level 1/3 stage (when ThYH_BNum2 ≧ YH_BNum> ThYH_BNum1)
D + (background) = correction level 0 stage (when ThYH_BNum1 ≧ YH_BNum)
And decide.

That is, the system controller 107 determines a larger dynamic range expansion amount as the area of the whiteout region in the image is larger.
Note that the method for determining the whiteout region is not limited to the method using the cumulative histogram described here, and any other method can be used.

  In step S208, the system controller 107 determines a final dynamic range expansion amount. Here, when the face detection is successful, the system controller 107 as the dynamic range expansion amount determination unit compares the dynamic range expansion amounts determined in S205 and S207 to determine the final dynamic range expansion amount. . For example, the system controller 107 can determine the larger dynamic range expansion amount as the final dynamic range expansion amount among the dynamic range expansion amount of the face area and the dynamic range expansion amount of the entire image.

  Alternatively, one of the dynamic range expansion amount of the face area and the dynamic range expansion amount of the entire image is determined as the final dynamic range expansion amount according to the imaging mode set by the mode dial included in the operation unit 112. Also good. For example, in the case of a human imaging mode (for example, portrait mode), the dynamic range expansion amount of the face area can be determined as the final dynamic range expansion amount. In the landscape imaging mode (for example, landscape mode), the dynamic range expansion amount of the entire image or the background area can be determined as the final dynamic range expansion amount.

  Further, the dynamic range expansion amount may be controlled according to the scene determination result. For example, at the time of backlight discrimination, the dynamic range expansion amount of the background area may be expanded to 2 steps (6/3 steps), and otherwise, it may be stopped to 1 step (3/3 steps).

  Further, a method other than selecting one of the dynamic range expansion amount of the face area and the dynamic range expansion amount of the entire image and determining the final dynamic range expansion amount may be employed. For example, the final dynamic range expansion amount can be determined according to the dynamic range expansion amount of the face area and the dynamic range expansion amount of the entire image.

FIG. 10 is a diagram illustrating a specific example of the dynamic range expansion amount finally determined according to the dynamic range expansion amount of the face area and the dynamic range expansion amount of the entire image.
In the examples shown in FIGS. 10A to 10C, the value of the dynamic range expansion amount finally determined is changed for each imaging mode. For example, when the dynamic range expansion amount of the face area is 1/3 step and the dynamic range expansion amount of the entire image is 0/3 step, the final dynamic range expansion amount is
Auto mode (FIG. 10A): 1/3 stage Portrait mode (FIG. 10C): 1/3 stage
Landscape mode (FIG. 10B): 0/3 stage.

In the example shown in FIGS. 10A to 10C, the dynamic range expansion amount of the background region is up to one stage (3/3 stage).
FIG. 10D shows an example in which the range of the dynamic range expansion amount finally determined according to the scene determination result is changed. In this example, when it is determined that the scene is a backlight scene, the dynamic range expansion amount of the face area is unchanged up to 3/3 step, but the dynamic range expansion amount of the background region is expanded to 6/3 step. Has been.

  The system controller 107 stores the dynamic range expansion amount determined as described above in the buffer memory 103, and refers to it during actual imaging. The determination operation of the dynamic range expansion amount can be performed at the time of standby, for example, every fixed frame or every fixed time of the EVF image, and the buffer memory 103 stores the latest dynamic range expansion amount.

(Scene judgment)
Next, a scene determination operation in the imaging apparatus according to the present embodiment will be described.
Although there is no particular limitation on the scene determined by the imaging apparatus of the present embodiment, here, as an example, the determination operation of presence / absence of a person, night scene, blue sky scene, sunset scene, backlight scene, and macro shooting will be described.
In this embodiment, the scene determination is
Exposure information such as subject brightness (Bv value) obtained from a photometric sensor or a captured image;
・ Subject distance information obtained by autofocus operation,
・ Color temperature information obtained by white balance control,
A color evaluation value for each block obtained by the color evaluation value acquisition circuit 132;
A histogram obtained by the histogram creation circuit 130,
Image sensor attitude information obtained from the output of the attitude sensor 133;
Based on the above, the system controller 107 executes.

(Night scene judgment)
When all of the following conditions (A1) to (A3) are satisfied, the night scene is determined.
(A1) Bv value is 0 [EV] or less (A2) The average luminance value of the entire image is less than or equal to a predetermined value calculated from the color evaluation value for each block (A3) A color within a predetermined range from the top of the image The average brightness obtained from the evaluation value and the average saturation are present at a predetermined ratio or more. The average brightness is the predetermined value Y_Th_night or less. The average saturation is the predetermined value C_Th_night or less. A determination can be made based on the output.
Further, the condition may be varied depending on the range of the Bv value.
Further, whether or not a tripod is attached may be considered. Whether or not the tripod is mounted can be determined from the output of the vibration detection sensor when the imaging apparatus includes a vibration detection sensor for correcting camera shake. Alternatively, the determination may be made based on the movement of the subject in the captured image. When it is determined that the tripod is mounted, it may be determined that the tripod night scene is different from the normal night scene.
For example, when there is no person on a tripod night view, there is less concern about camera shake and subject blur, so exposure control may be performed such that exposure is reduced for a long time with reduced sensitivity.
In addition, when there is a person on a tripod night view, it is desirable to control exposure and flash light control so that the brightness of the face area and the background is optimized.
In addition, exposure to long seconds makes it easier for black to float and noise to increase. Therefore, it is possible to obtain a more favorable image by reducing the gradation characteristics of dark parts (suppressing the gradation level to a low level). It is done.

(Blue sky judgment)
When all of the following conditions (B1) to (B4) are satisfied, it is determined as a blue sky.
(B1) Bv value is 5 [EV] or more (B2) Subject distance at the time of shooting is a predetermined value or more (not macro shooting)
(B3) The color temperature is in a predetermined color temperature range (B4) A block satisfying the following conditions in the average luminance, average hue, and average saturation obtained from the color evaluation value is within a predetermined range from the top of the image. Exceeded percentage The average brightness is within the specified range (Y_Th_Sky_Low or higher, Y_Th_Sky_Hi or lower)
Average hue is within the specified range (Hue_Th_Sky_Low or higher, Hue_Th_Sky_Hi or higher)
Average saturation is within the specified range (C_Th_Sky_Low or higher, C_Th_Sky_Hi or lower)
Each condition may be varied according to the range of the Bv value.
For example, when there is no person in the blue sky, it is considered that the probability of being a landscape photograph is high. For this reason, it is possible to obtain a more preferable image by performing a process of enhancing the overall saturation of blue and other colors and increasing the contrast.
If there is a person in the blue sky, there is a high probability that it is a landscape snapshot. Therefore, it is possible to obtain a preferable image by emphasizing the saturation in the vicinity of the skin color while emphasizing the saturation of blue, and performing the AE and the gradation characteristic setting so that the exposure and gradation of the face are optimized.

(Evening judgment)
When all of the following conditions (C1) to (C7) are satisfied, the scene is determined to be an evening scene.
(C1) Bv value is 7 [EV] or more (C2) The subject distance at the time of shooting is a predetermined value or more (not macro shooting)
(C3) The average luminance value of the entire image calculated from the color evaluation value is less than or equal to a predetermined value (C4) There are more than a predetermined ratio of high luminance blocks whose average luminance is a predetermined value or more (C5) Obtained from the color evaluation value Blocks that satisfy the following conditions for average luminance, average hue, and average saturation exist in a predetermined ratio or more: Average luminance is within a predetermined range (Y_Th_Sunset_Low or higher, Y_Th_Sunset_Hi or lower)
Average hue is within the specified range (Hue_Th_Sunset_Low or higher, Hue_Th_Sunset_Hi or lower)
Average saturation is within the specified range (C_Th_Sunset_Low or higher, C_Th_Sunset_Hi or lower)
(C6) The number of blocks that do not satisfy any of the following conditions is equal to or less than a predetermined value. The condition of (C5) for evening scene determination. The condition of (B4) for blue sky determination. The average saturation is equal to or less than a predetermined value. (High brightness block)
Average brightness is below the specified value (low brightness block)
(C7) The hue or saturation histogram in the color evaluation value has a variance greater than or equal to a predetermined value. Each condition may be varied according to the range of the Bv value. The smear area may be removed based on the color evaluation value.
If it is determined that the scene is an evening scene, a more preferable image can be obtained by performing saturation enhancement of red / orange colors, WB setting in a high color temperature direction such as cloudiness, and performing AE control with slightly underexposure.

(Backlight judgment)
(1) When a face is detected When the difference between the average luminance value obtained from the color evaluation value of the block corresponding to the face area and the average luminance value obtained from the color evaluation value of another block is equal to or greater than the predetermined EV value It is determined that there is a person and backlight.
(2) When no face is detected When AE converges for a certain period, as shown in FIG. 15, the entire image is divided into four types of areas A to D in order from the outer periphery, and blocks corresponding to the respective areas. The average luminance value for each region is calculated from the color evaluation value.
The average luminance value of the area A is compared with the average luminance values of the areas BD, and when a difference equal to or greater than a predetermined EV value is detected, it is determined that there is no person backlight.
In addition to the method described here, luminance information obtained by dividing a CCDRAW image into blocks may be used, or histogram information may be used.
The backlight determination in the present embodiment detects a two-dimensional pattern of brightness, and is different from the detection of the degree of brightness whiteout for each block or area, as performed in the detection of whiteout. Therefore, the backlight determination and the whiteout detection are different processes.
However, scenes that have been determined to be backlit are considered to have a high possibility of gradation loss in high-luminance portions, so if a backlit determination is made, a more preferable image can be obtained by setting a larger dynamic range expansion amount. Obtainable.

(Macro judgment)
Whether it is macro photography can be determined based on the focused subject distance. The subject distance can be obtained by referring to a preset correspondence table from the position of the focus lens, for example. When the subject distance is equal to or smaller than a predetermined value, it is determined that the macro photography is performed.
If the flash photography is performed at a short distance, there is a high possibility of overexposure. Therefore, it is desirable to control the exposure so that the flash is not emitted as much as possible when it is determined to be macro photography. In addition, when the flash photography is performed, the tone processing characteristic in which the high luminance part is laid down is used in the signal processing circuit 140 so that the gradation of the highlight part is not easily saturated.

FIG. 14 is a flowchart for explaining the operation of the main imaging process in the imaging apparatus of the present embodiment.
Note that, at the time of imaging standby, the system controller 107 determines the dynamic range expansion amount from the EVF image by the above-described method, for example, at a constant cycle. In this embodiment, it is assumed that the dynamic range expansion amount (AE target value reduction amount) can be determined in four steps from 0/3 step to 3/3 step in units of 1/3 step. The range of the dynamic range expansion amount and the size per stage can be arbitrarily set. Further, for example, when the shutter button included in the operation unit 112 is pressed halfway and an imaging preparation instruction is input, the system controller 107 uses the EVF image and the results of AF control and AE control to perform the scene determination process described above. Execute.

  Then, at the time of imaging standby, in response to the shutter button included in the operation unit 112 being fully pressed and an imaging start instruction being input, the system controller 107 as imaging control means starts the following processing.

  In step S <b> 400, the system controller 107 determines whether it is determined as a sunset scene as a result of the scene determination. If it is determined that the scene is an evening scene, the system controller 107 calculates an exposure correction amount for the evening scene in S408, and does not calculate the dynamic range expansion amount. On the other hand, if it is not determined to be a sunset scene, the system controller 107 acquires the dynamic range expansion amount determined immediately before the imaging start instruction is input from the buffer memory 103 in S401.

  In step S <b> 402, the system controller 107 determines whether the acquired dynamic range expansion amount, that is, the reduction amount of the AE target value can be realized by sensitivity adjustment (control of the CDS gain circuit and the VGA gain circuit) in the AFE circuit 150. . This determination can be made by comparing the sensitivity range adjustable by the AFE circuit 150 with the dynamic range expansion amount acquired in S401. If the sensitivity corresponding to the dynamic range expansion amount cannot be reduced (gain reduction), the system controller 107 determines that the dynamic range expansion amount cannot be realized only by sensitivity adjustment in the AFE circuit 150.

  When the dynamic range expansion amount can be realized by sensitivity adjustment in the AFE circuit 150, the system controller 107 as the correction amount determination unit calculates the gain setting as an imaging condition in S404. Note that there is no particular limitation on how the CDS gain and the VGA gain are set in combination, and any combination can be set.

On the other hand, when it is determined that the dynamic range expansion amount cannot be realized only by the sensitivity adjustment in the AFE circuit 150, the system controller 107 performs exposure based on the gain amount that is still insufficient even if the AFE circuit 150 performs the gain control that is possible. The condition is changed (S405). Specifically, the system controller 107 calculates an exposure correction amount for realizing an insufficient gain amount.
The exposure correction here is a minus correction, and can be realized by a general method such as reducing the aperture, increasing the shutter speed, or inserting a neutral density filter (ND filter).

  In step S406, the system controller 107 determines whether the exposure correction calculated in step S405 is possible. For example, in an imaging apparatus that does not have an ND filter, the exposure cannot be negatively corrected when the shutter speed is already set to the maximum speed and the aperture is set to the minimum (the aperture value is maximum) by automatic exposure control. Also, when the maximum shutter speed that can be tuned is set during flash imaging, the shutter speed cannot be increased. The same applies to the case where the upper limit of the shutter speed is determined. In the aperture priority AE mode, it is preferable that the aperture value set by the user is not changed. Therefore, if the shutter speed is already the highest speed, it may be determined that minus correction cannot be performed. The same applies to the shutter speed priority AE mode.

  If it is determined that the negative exposure correction corresponding to the insufficient gain adjustment cannot be performed, the system controller 107 corrects the dynamic range expansion amount to the maximum value that can be realized by sensitivity adjustment and exposure correction in S407. . Then, the system controller 107 calculates a gain amount set in the AFE circuit 150 and an exposure correction amount as necessary.

  In step S409, the system controller 107 sets a gain amount as an imaging condition in the CDS gain circuit 152 and the VGA gain circuit 154 of the AFE circuit 150. When performing exposure correction, the exposure parameters (shutter speed, aperture value, use / non-use filter setting, etc.) corresponding to the AE result are changed according to the exposure correction amount, and the lens apparatus is also used as the imaging condition. Set to 200.

In step S410, the system controller 107 performs still image shooting (main exposure).
In step S411, the system controller 107 determines whether the dynamic range has been expanded in the process before the main exposure. Here, the system controller 107 determines that the dynamic range has not been expanded when the amount of expansion of the dynamic range is zero and when it is determined to be a sunset scene in S400.
In step S412, the system controller 107 determines whether the scene is determined to be a night view as a result of the scene determination process before the main exposure. If it is determined that the scene is a night scene, the gradation characteristic for night scene is selected according to the dynamic range expansion amount in S415. If the night scene is not determined, a histogram is calculated from the main exposure image in S413, and the reduction amount (D-amount) of the dynamic range is determined in S414.

FIG. 16 shows an example of gradation characteristics to be selected when the night scene is determined. As shown in the figure, noise in the dark part can be made inconspicuous by crushing the dark part characteristic (lowering the increase rate of the output luminance with respect to the increase rate of the input luminance value) compared to the normal gradation characteristic. Therefore, a preferable image with black tightened is obtained as an image at the time of night scene photography.
Of course, these gradation characteristics may be varied depending on sensitivity, aperture, zoom position, and the like. For example, since the amount of light around the lens tends to drop when the aperture is open, the degree of black loss may be reduced.

In S416, the system controller 107
-Tone characteristics according to the dynamic range expansion amount (except when the expansion amount is 0 or the evening scene is judged),
Tone characteristics for night scene selected in S415 or Tone characteristics according to the dynamic range reduction amount determined in S414,
Are set in the signal processing circuit 140.

FIG. 11 is a diagram illustrating a conceptual diagram of a dynamic range in the present embodiment.
In the present embodiment, the dynamic range is defined as the ratio of the saturation signal amount luminance of the image sensor to the appropriate luminance. The appropriate luminance is a luminance target value level when performing automatic exposure control (AE). For example, if the AE mode is the average photometry mode, it corresponds to the average value of the screen luminance.

Therefore,
Dynamic range = sensor saturation signal amount luminance / AE target value.
Here, the AE target value is an AE target value based on the output signal of the image sensor 101 before sensitivity adjustment is performed by the AFE circuit 150.

The AE target value may change according to the AE mode, and the AE target value in each mode can be used even in the evaluation photometry mode or the spot photometry mode.
FIG. 12 is a diagram illustrating a relationship example between the AE target value, the saturation signal value, and the dynamic range.
FIG. 12 shows that the dynamic range amount can be increased by lowering the AE target value.

FIG. 13 is a diagram illustrating an example of setting gradation characteristics in the signal processing circuit 140 according to the present embodiment.
Setting example of gradation characteristics (brightness correction amount) when the dynamic range expansion amount is set to four levels of normal (0/3 step), +1/3 step, +2/3 step, and +3/3 step Show.

  Here, the AE target value corresponding to each dynamic range expansion amount is the same as that shown in FIG. As shown in FIG. 13, the AE target value after gradation correction is performed on the AE target value at the time of each dynamic range expansion becomes a normal AE target value that does not expand the dynamic range regardless of the dynamic range expansion amount. The gradation characteristics are set as follows.

  As described with reference to FIGS. 11 and 12, the dynamic range can be expanded by lowering the AE target value. However, simply lowering the AE target value results in underexposure and the captured image becomes dark. Therefore, the signal processing circuit 140 performs gradation correction so that the image data after imaging is brightened according to the dynamic range expansion amount, thereby expanding the dynamic range while making the brightness (exposure) of the captured image appropriate. be able to. Therefore, the signal processing circuit 140 operates as correction means.

  In the present embodiment, the configuration in which the decrease in luminance of the captured image due to the decrease in the AE target value is compensated by gradation correction is described. May be.

  Further, a gain such as a gain of a white balance coefficient for correcting white balance and a clipping amount for determining a saturation signal amount may be controlled. That is, after the A / D conversion is performed on the image signal whose gain has been reduced by reducing the exposure amount or the AFE gain, the gain is increased by the signal processing circuit in the subsequent stage, and the clipping amount is increased by the gain increase (saturation signal amount The same effect can be obtained.

Here, the relationship between sensitivity setting (gain setting of the AFE circuit 150) and saturation of the image sensor 101 will be described.
Generally, the gain for the output signal of the image sensor is controlled so that the output satisfies a predetermined relationship with respect to the light amount and the exposure value (input) of the camera.

  However, there is an upper limit on the amount of charge that can be accumulated in the photodiode of the image sensor. Therefore, if the gain applied by the AFE circuit 150 is reduced for the purpose of reducing the imaging sensitivity of the image sensor, the maximum signal amount after gain application is also reduced. Therefore, the saturation signal amount decreases as the gain decreases.

  As described above, in the direction of increasing the sensitivity of the image sensor, it is possible to set a desired sensitivity if the amplification of noise is ignored, whereas in the direction of decreasing the sensitivity, the limit caused by the saturation signal amount. Value exists.

  In S408 of FIG. 14, when the sensitivity cannot be lowered, the minimum sensitivity that can be set by the imaging apparatus is often set. This means that the gain for the output signal of the image sensor has already been reduced to a value corresponding to the minimum sensitivity. Therefore, the sensitivity cannot be further reduced by controlling the gain of the AFE circuit 150. Therefore, in this embodiment, when the target AE target value (sensitivity) reduction amount cannot be realized even if the gain of the AFE circuit 150 is controlled, further sensitivity reduction is realized by exposure correction.

  The operation of performing imaging with reduced sensitivity by exposure correction and correcting the brightness by correcting the gradation of the obtained dark image by the signal processing circuit 140 is the same as increasing the sensitivity after all, Noise is also amplified at the time of gradation correction, resulting in degradation of image quality.

  However, in the present embodiment, when the reduction of the imaging sensitivity corresponding to the dynamic range expansion amount can be realized by the gain control of the AFE circuit 150, the imaging sensitivity reduction by the gain control is performed with priority. Even when the gain control alone cannot achieve a reduction in imaging sensitivity corresponding to the dynamic range expansion amount, the gain is reduced to the minimum, and the insufficient sensitivity reduction is compensated by exposure correction. In this case, even if noise is amplified during gradation correction, the noise itself is not large because the gain of the AFE circuit 150 has already been reduced to a level corresponding to the lowest sensitivity. For this reason, deterioration in image quality can be minimized.

FIG. 17 is a flowchart for explaining the dynamic range reduction amount (D−) determination operation in the imaging apparatus of the present embodiment. The operation shown in FIG. 17 corresponds to the histogram acquisition in S413 and the dynamic range reduction amount determination operation in S414 in FIG.
In step S <b> 501, the system controller 107 multiplies the CCD RAW image data stored in the buffer memory 103 by a white balance coefficient (WB coefficient) for each of R, G, and B colors. The WB coefficient used here may be a WB coefficient calculated from an actually captured image or a WB coefficient calculated from an EVF image most recently captured. However, in the latter case, when flash photography is performed with actual imaging, a WB coefficient corresponding to the color temperature of the flash light or a WB coefficient for a D55 light source or daylight can be used.
By multiplying by the WB coefficient, it is possible to accurately detect the saturation amount for each color, and it is possible to detect even a subject in which a specific color is saturated.

In step S <b> 502, the system controller 107 uses the histogram creation circuit 130 to obtain a cumulative histogram for each color related to the image multiplied by the WB coefficient.
In step S503, the system controller 107 acquires a luminance value Y_Hp for which the frequency of the cumulative histogram is a predetermined value (for example, 80%) for each color as a highlight luminance value.
In step S504, the system controller 107 calculates a gain value Gain_yhp for a predetermined saturation target value Sat_TargetY for each color using the following equation.
Gain_yhp = Sat_TargetY / Y_hp
Then, the system controller 107 sets the minimum gain value in the Gain_yhp calculated for each color as the gain value Gain_yhp corresponding to the final dynamic range reduction amount.

In step S505, the system controller 107 selects a gradation characteristic parameter corresponding to the gain value calculated in step S504.
FIG. 18 is a diagram illustrating a relationship example between the final gain value Gain_yhp calculated in S504 and the final gain increase amount according to the imaging sensitivity.
In the present embodiment, the final gain increase amount is set to four types of 0, 1.12, 1.25, and 1.41, and the system controller 107 is
1.12-> Tone characteristics corresponding to 1/6 step up exposure-> Tone characteristics corresponding to 1/3 step up exposure-1.41-> Tone characteristics corresponding to 1/2 step up exposure select.

  FIG. 19 is a diagram illustrating the relationship between the gain increase amount and the gradation characteristic setting value. Increasing the exposure by 1/6 is the same as increasing the AE target value by 1/6. Thus, the dynamic range reduction process corrects the gradation characteristics of the image obtained by the main exposure, and brightens the image by the number of steps corresponding to the final gain increase amount.

In this embodiment, the dynamic range reduction amount is changed by changing the gradation characteristic setting. However, the dynamic range reduction amount can be changed by using a lookup table or by controlling the WB coefficient.
Further, in the present embodiment, histogram analysis is performed on the RAW image at the time of the main exposure, and the dynamic range reduction amount is determined. However, similarly to the determination of the dynamic range expansion amount, exposure control may be performed by performing histogram analysis on the EVF image and determining the dynamic range reduction amount. In this case, since it is not necessary to increase the gain according to the gradation characteristics, it is possible to reduce the dynamic range without accompanying noise amplification due to gain increase.
However, it may not be possible to determine the degree of whiteout of an image until the main exposure is performed, for example, during flash photography. Therefore, for example, when the flash is not emitted, the dynamic range reduction amount is determined based on the EVF image to perform exposure control, and when the flash is emitted, the analysis result of the main exposure image may be used.

  FIG. 20 shows an example of the relationship between exposure control, dynamic range expansion processing (D +), dynamic range reduction processing (D−), and night scene tone characteristics setting according to the scene determination result in the imaging apparatus of the present embodiment. FIG. In the present embodiment, the dynamic range reduction process is performed when the dynamic range is not expanded. Therefore, in FIG. 20, the fact that ○ (or ◎) is written in both D + and D- indicates that D + or D- can be executed, and that both are performed. Absent.

When the scene is determined to be a night view and there is no person (when no face is detected by the face detection circuit 120), dynamic range expansion processing is performed according to the dynamic range expansion amount determined based on the EVF image. In addition, as described above, the gradation characteristic with the dark part crushed is selected.
On the other hand, in the case of a night scene, since the area of a dark part is large, when dynamic range reduction processing is performed using luminance information such as a histogram, gradation characteristics with increased sensitivity are selected. However, since it is preferable that the night view is photographed to a certain degree of darkness compared to a normal scene, the dynamic range reduction process (D−) is not performed.

On the other hand, when it is determined that the scene is a night scene and a person is included, the dynamic range is expanded or reduced, and the gradation characteristics for the night scene are not selected.
When a night scene includes a person, if the gradation characteristics for night scene are selected, the contrast of the face increases too much, and the gradation of the dark part of the face is crushed so that a preferable image cannot be acquired. In addition, when a person is included in a night view, flash photography is often performed. However, when the subject distance is long, such as when photographing is performed on the telephoto side of the zoom lens, the flash light often does not reach sufficiently. Therefore, it is possible to obtain an image that is preferably subjected to the dynamic range reduction process for a human face region or the like.

If it is determined that the scene is a sunset scene and no person is included, exposure control is performed so that the exposure is underexposed compared with normal shooting in main exposure, and dynamic range expansion processing is prohibited.
In many scenes that are determined to be sunset scenes, the sun is often included in the angle of view. For this reason, since the result of automatic exposure control tends to be underexposed, there is little effect of further correcting the minus and expanding the dynamic range. Rather, the dynamic range expansion process (D +) may increase the imaging sensitivity, and in that case, there is a high possibility that noise is amplified and an undesirable image is obtained. In addition, the saturation is controlled so as to emphasize red.

Also, when a person is included in the sunset and the flash is not emitted, the brightness of the face cannot be controlled. Therefore, it is possible to obtain a preferable image by performing exposure determination with face priority (face priority AE) and performing dynamic range expansion processing (D +) or reduction processing (D−).
On the other hand, when a person is included in the evening scene and the flash is emitted, the brightness of the face can be controlled by the flash light control. Therefore, as in the case where there is no person, it is possible to obtain a preferable image by capturing an image with underexposure and prohibiting the dynamic range expansion process. In addition, when a person is included in the evening scene, red is not emphasized again so as not to affect the tone of the skin color.

If the scene is determined to be backlit and no person is included, the scene is often sufficiently bright. That is, the imaging sensitivity is low and a high shutter speed is often set by the imaging device. As a result, there is a high possibility that the dynamic range expansion process cannot be performed. In the case of imaging in which a person is not included and the landscape is the main subject, an image with less noise is often preferred.
On the other hand, when a person is included, the person is often in the shade or in a dark place in the room and the background is in the sun or outdoors. Thus, when different light sources are mixed in one shooting scene, the dynamic range expansion amount is often insufficient in one stage. Further, in the case of a snapshot, it is preferable that the face and background are taken.

In this way, in a backlight scene, if the person is included, the dynamic range expansion amount is increased, and if the person is not included, the dynamic range expansion amount is not increased so that a more preferable image can be taken. Become. For example, in a normal scene determination result, the dynamic range expansion amount is set to a maximum of one level, and in the case of a scene that includes a person in backlight and is expanded, the dynamic range expansion amount is expanded to, for example, two levels.
Of course, it goes without saying that the upper limit of the enlargement amount determined in the dynamic range enlargement process may be determined according to the degree to which the imaging sensitivity is increased. For example, it is possible to control so that the dynamic range expansion amount is increased in the case of brightness at high sensitivity, and otherwise the dynamic range expansion amount is decreased.

When it is determined that the sky is blue, it is the same as the normal scene except that the saturation of blue is emphasized.
However, when it is determined that the sky is blue and the luminance of the blue portion is high and there are many blocks in the cyan hue direction, the upper limit of the dynamic range expansion amount may be extended. As a result, it is possible to determine the dynamic range expansion amount while positively determining the degree of jump of the empty portion.
In the case of a person scene, AE with priority on person is performed and dynamic range expansion or reduction processing is performed.

Normal scenes (when night scenes, evening scenes, backlights are not detected, and no person is detected), dynamic range expansion processing is performed within the range where the imaging sensitivity does not change, and dynamic range reduction processing is performed only at low imaging sensitivity, and noise It is possible to suppress the increase in
When it is determined that there is no blue sky (no person), it is the same as the normal scene except that the blue saturation is emphasized.
However, when it is determined that the sky is blue and the luminance of the blue portion is high and there are many blocks in the cyan hue direction, the upper limit of the dynamic range expansion amount may be extended. As a result, it is possible to determine the dynamic range expansion amount while positively determining the degree of jump of the empty portion.

(Other embodiments)
In the above-described embodiment, the case where the dynamic range expansion according to the present invention is applied to still image capturing has been described. However, the present invention can be similarly applied to EVF image capturing and moving image capturing. In this case, the timing for setting these parameters is adjusted so that the gain control of the AFE circuit 150 (and exposure correction as necessary) and the gradation correction in the signal processing circuit 140 are applied to the same image.

Claims (8)

Scene determination means for performing scene determination based on a moving image captured during standby of still image capturing;
Control means for controlling the imaging operation,
Based on the result of the scene determination,
A dynamic range expansion process that captures a still image with reduced imaging sensitivity, and applies gradation correction to compensate for the reduction in the imaging sensitivity with respect to the captured image;
A control unit that captures a still image and executes dynamic range reduction processing that applies gradation correction based on the moving image or the captured image to the captured image. A characteristic imaging device.   The imaging apparatus according to claim 1, wherein the scene determination unit determines at least one of a presence / absence of a person, a night scene, a blue sky scene, an evening scene, a backlight scene, and macro photography.   The said control means performs the said dynamic range expansion process, when the result of the said scene determination is a predetermined scene which does not perform the said dynamic range expansion process, The said dynamic range expansion process is performed. Imaging device.   The control means reduces the dynamic range when the scene determination result is a scene for which the dynamic range expansion process is not performed and when the imaging sensitivity reduction amount in the dynamic range expansion process is zero. The imaging apparatus according to claim 1, wherein processing is executed.   5. The control unit according to claim 1, wherein, in the dynamic range expansion process, the reduction amount of the imaging sensitivity is determined to be larger as the whiteout amount of the image subjected to the scene determination is larger. The imaging device according to any one of the above.   In the dynamic range reduction process, the control unit is configured such that a luminance value at which a frequency of a cumulative histogram of an image captured in the still image has a predetermined value corresponds to a saturation signal amount of an imaging element included in the imaging device. 6. The imaging apparatus according to claim 1, wherein gradation correction is applied to the still image captured image. A scene determination step for determining a scene based on a moving image captured at the time of standby for still image capturing;
A control method for an image pickup apparatus having a control process for controlling an image pickup operation,
The control step is
Based on the result of the scene determination in the scene determination step,
A dynamic range expansion process that captures a still image with reduced imaging sensitivity, and applies gradation correction to compensate for the reduction in the imaging sensitivity with respect to the captured image;
An imaging apparatus that captures a still image and performs dynamic range reduction processing that applies gradation correction based on the moving image or the captured image to the captured still image Control method.   The program for making a computer of an imaging device perform each process of the control method of the imaging device of Claim 7.