JP5929019B2 - Imaging device - Google Patents

Description

The present invention relates to an imaging equipment.

  Conventionally, when shooting a scene with a large luminance difference, the image was processed so that the data in the bright part was not saturated by lowering the exposure, and then the gain in the dark part was increased to make it brighter, thereby increasing the dynamic range of the image. A method of enlarging is known (see, for example, Patent Document 1).

JP 2010-232711 A

  However, in the above conventional technique, if the exposure is greatly lowered when the face is dark like a backlit person, the S / N of the data of the face portion of the photographed image may deteriorate, and the face may not be detected. is there. Even if it can be detected, noise may become conspicuous if the gain of the face portion data is increased.

An object of the present invention, face detection can be face even when dark, it is to provide an imaging equipment capable and noise of obtaining an image as inconspicuous.

The present invention is, that solve the problems by following such a solution.

The present invention is based on a photometry unit that divides a subject field into a plurality of areas and performs photometry, a first exposure calculation unit that calculates a first exposure value based on a photometry result by the photometry unit, and the photometry result A correction amount calculation unit that calculates a correction amount for restoring the bright portion of the subject with respect to the first exposure value, a main subject detection unit that detects a main subject based on the photometric result, A main subject luminance calculation unit for calculating luminance; a correction amount limiting unit for limiting the correction amount based on the luminance of the main subject; and a first exposure value corrected for the first exposure value based on the limited correction amount. A second exposure calculation unit that calculates an exposure value of 2, and an imaging unit that images a subject based on the second exposure value and generates image data, and the correction amount is equal to or greater than a limit value, The limited correction amount is the limit value, and the limit value The minus under allowance of the face from the first exposure value, Ru imaging apparatus der a value obtained by subtracting from the representative luminance of the main object.
In addition, the present invention provides a first exposure calculation unit that calculates a first exposure value based on a photometry result obtained by a photometry unit that performs photometry by dividing an object scene into a plurality of areas, and a first exposure based on the photometry result. A first calculation unit that calculates a correction amount for restoring the bright portion of the subject with respect to the exposure value; a second calculation unit that calculates a limit value for limiting the correction amount based on the luminance of the main subject; A selection unit that selects the limit value when the limit value is smaller than the correction amount, and the first exposure value based on the limit value selected by the selection unit when the limit value is smaller than the correction amount. A second exposure calculation unit that calculates the corrected second exposure value, an imaging unit that captures an image of a subject based on the second exposure value and generates image data, and a gradation so that an image of the image data becomes brighter An image processing unit that corrects, and the limit value is The minus the under-capacity of the face from the serial first exposure value, the an image pickup apparatus which is a value obtained by subtracting from the luminance of the main object.

According to the present invention, can be detected face even when the face is dark, it is possible to provide an imaging equipment to and capable of noise to obtain an image as inconspicuous.

It is a figure which shows the structure of the camera of embodiment. It is a functional block diagram of the camera of the embodiment It is a flowchart which shows the basic operation | movement of the camera of embodiment. It is a figure which shows the characteristic of a low-pass filter. It is the figure which illustrated the gain improvement function corresponding to the brightness improvement amount. It is a flowchart which shows a photometry calculation subroutine. It is a figure which shows a 2nd image sensor, (a) is an example of the pixel arrangement | sequence of a 2nd image sensor, (b) is a pixel divided | segmented into 24 area | regions, (c) is an example of an imaging | photography scene, (d) is a face of a face. It is a figure which shows a center part. It is a flowchart which shows the specific calculation method of an exposure control value. (A) is a figure which shows the luminance value of each area | region divided into 24, (b) is the figure which showed the example which put 24 divisions into 4 each. It is a flowchart which shows the calculation of temporary highlight restoration amount. It is a figure which shows the selection method of the open area | region of a main to-be-photographed object. It is a figure showing an example of division of a main subject. It is a figure which shows the example of calculation in the case of a variable.

Embodiments of the present invention will be described below.
FIG. 1 is a diagram illustrating a configuration of a camera 100 according to the present embodiment.
The camera 100 includes a camera body 1 and a lens unit 2. Here, the lens unit 2 is attached to the camera body 1 in a replaceable manner. The lens unit 2 includes a photographing lens 3, a motor 4 that drives the photographing lens 3, and an in-lens memory 18 that records information on the lens unit.

  The camera body 1 includes a main mirror 5, a sub mirror 6, a focal plane shutter 8, a first image sensor 9 that photoelectrically converts a subject image, a finder optical system 20, and a second image sensor 16 that photoelectrically converts the subject image. And a focus detection unit 7, a control unit 10, a camera shake sensor 17 (for example, an angular velocity sensor), and a display unit 19.

  Here, the first imaging sensor 9 is a sensor for generating a high-quality still image. The second imaging sensor 16 is a sensor for analyzing a shooting scene and calculating shooting and image processing conditions of the first imaging sensor 9. It has enough pixels to analyze the shooting scene. FIG. 7A described later is an example of a pixel array of the second image sensor 16.

  The main mirror 5, the sub mirror 6, the focal plane shutter 8, and the first image sensor 9 are arranged along the optical axis of the photographing lens 3, the finder optical system 20 is in the upper region of the main mirror 5, and the focus detection unit 7 is the sub mirror. 6 is arranged in the lower region.

Here, when the photographer observes the object scene through the finder optical system 20, that is, other than still image shooting and live view, the main mirror 5 has a shooting optical path from the shooting lens 3 to the first image sensor 9. The light beams that intersect and pass through the photographing lens 3 are reflected upward and guided to the finder optical system 20 (solid line position).
On the other hand, at the time of live view and still image shooting, the main mirror 5 is flipped upward, and the light flux that has passed through the shooting lens 3 is guided to the focal plane shutter 8 and the first image sensor 9, and the image of the first image sensor 9 Based on the signal, photographic image data of the subject can be generated (position of dotted line).

  Further, the central portion of the main mirror 5 is a half mirror that transmits light, and a part of the light beam that has transmitted through the main mirror 5 during non-shooting is reflected downward by the sub mirror 6 and guided to the focus detection unit 7. It is like that.

  The finder optical system 20 includes a diffusing screen 11, a condenser lens 12, a pentaprism 13, an eyepiece lens 14, and a re-imaging lens 15.

  The diffusing screen 11 is disposed above the main mirror 5 and once forms an image of the light beam reflected by the main mirror 5 at the position of the solid line. The light beam formed on the diffusion screen 11 passes through the condenser lens 12, the pentaprism 13, and the eyepiece lens 14 and reaches the eyes of the photographer. A part of the light beam diffused by the diffusing screen 11 passes through the condenser lens 12, the pentaprism 13, and the re-imaging lens 15 and reaches the second imaging sensor 16.

  FIG. 2 is a functional block diagram of the camera 100 of the embodiment. As shown in FIG. 2, the camera 100 includes a timing generator 20, a signal processing unit 21, an A / D conversion unit 22, a buffer memory 23, a bus 24, a card interface 25, and a compression / decompression unit 26 in addition to the configuration of FIG. Each part of the image display unit 27 is provided. The timing generator 20 supplies output pulses to the first image sensor 9. Further, the image data generated by the first imaging sensor 9 is temporarily stored in the buffer memory 23 via the signal processing unit 21 (including a gain adjustment unit corresponding to the imaging sensitivity) and the A / D conversion unit 22. . The buffer memory 23 is connected to the bus 24. The bus 24 is connected to the card interface 25, the control unit 10, the compression / decompression unit 26, and the image display unit 27 described in FIG. The card interface 25 is connected to a removable memory card 28 and records image data on the memory card 28.

  The control unit 10 includes an exposure value calculation unit 10a, a restoration amount calculation unit 10b, a main subject detection unit 10c, a main subject luminance calculation unit 10d, and a restoration amount limit unit 10e, which will be described later. The control unit 10 is connected to a switch group 29 (including a release button (not shown)) of the camera 100, a timing generator 20, and the second imaging sensor 16. Further, the image display unit 27 displays an image or the like on the display unit 19 provided on the back surface of the camera 100.

Next, the operation will be described based on the flowchart of FIG.
When a release button (not shown) is half-pressed, the control unit 10 calculates the exposure control value BvCntl1 by performing the operation of the photometric calculation subroutine of step 101 while keeping the main mirror 5 at the position of the solid line. Details will be described later.

  In step 102, it is determined whether or not the release button has been fully pressed. If not fully depressed, the process returns to step 101. If fully depressed, the process proceeds to step 103.

  In step 103, the control unit 10 controls each unit, and based on the result of the photometric calculation performed in step 101, the first image sensor 9 captures a subject image and generates image data. The image data generated by the first imaging sensor 9 is temporarily stored in the buffer memory 23 via the signal processing unit 21 and the A / D conversion unit 22.

  In step 104, the control unit 10 reads image data from the buffer memory 23 and performs face detection. Since the specific method of face detection is the same as that of the known technique, description thereof is omitted. When a plurality of persons are included in the detection target image data, the control unit 10 detects each person's face.

  In step 105, the control unit 10 performs normal image processing on the image data read from the buffer memory 23 in step 104. Normal image processing includes white balance adjustment, interpolation processing, color tone correction processing, gradation conversion processing, and the like. Since the specific method of each process is the same as that of a well-known technique, description is abbreviate | omitted.

  In step 106, the control unit 10 determines whether or not the gradation compression mode is set. If the control unit 10 determines that the tone compression mode is set, the process proceeds to step 107. On the other hand, if it is determined that the mode is not the gradation compression mode (the gradation non-compression mode), the control unit 10 proceeds to Step 111 described later.

In step 107, the control unit 10 creates a low-pass image from the image data read from the buffer memory 23 in step 104.
The control unit 10 creates a low-pass image Ly from the image data read from the buffer memory 23 in step 104 using the following equation.

  Y in Equation 1 indicates the luminance value of the target pixel. Further, kr, kg, and kb in Equation 1 are predetermined coefficients. Lpw in Equation 2 is a low-pass filter around the pixel of interest, and this low-pass filter has the characteristics shown in FIG. This low-pass filter is a wide-width low-pass filter whose half-value width (d in FIG. 4) is 1/100 or more of the short side of the image.

  In step 108, the control unit 10 calculates an evaluation value related to brightness. First, the control unit 10 obtains the center coordinates (Fx, Fy) of the face area based on the result of the face detection performed in step 104. However, when a plurality of faces are detected in step 104, the center coordinates of the face area having the largest area are obtained.

  Next, the control unit 10 obtains a facial brightness Yf as an evaluation value related to brightness. The brightness Yf of the face is obtained by the following equation.

Yf = Ly [Fx, Fy] (Formula 3)
In the case where a plurality of faces are detected in step 104, the example of obtaining the center coordinates of the face area having the largest area has been shown. However, the center coordinates of the darkest face area may be obtained. In this way, by performing the processing according to the darkest face region, all of the plurality of faces can be expressed with a preferable brightness.

Alternatively, the brightness Yf may be obtained for each of a plurality of faces, and weighting may be performed according to the area of each face region.
In step 109, the control unit 10 determines the brightness improvement amount when correcting the dark part gradation. First, the control unit 10 obtains a brightness improvement amount candidate fup according to the face brightness Yf obtained in step 108 based on Table 1.

  However, the control unit 10 limits the brightness improvement amount candidate fup according to the ISO sensitivity at the time of imaging performed in step 103. That is, the control unit 10 restricts the obtained brightness improvement amount candidate fup based on Table 2.

  Next, the control unit 10 obtains a brightness improvement amount candidate Hup according to the highlight restoration amount HiRcv obtained in step 206 of FIG.

  Then, the control unit 10 compares the brightness enhancement amount candidate fup according to the face brightness Yf with the brightness enhancement amount candidate Hup according to the highlight restoration amount HiRcv, and the larger value is used as the highlight restoration amount HiRcv. The brightness improvement amount Gup is set accordingly.

  In step 110, the control unit 10 performs gradation compression processing on the image data that has been subjected to normal image processing in step 105 in accordance with the brightness enhancement amount Gup determined in step 109.

First, the control unit 10 selects the gain improvement function fg according to the lightness improvement amount Gup determined in step 109.
FIG. 5 is a diagram showing a gain enhancement function fg for gradation compression corresponding to the brightness enhancement amount Gup. The gain improvement function fg has a gain corresponding to the luminance Y of the image. And the gain improvement function fg becomes large, so that the brightness | luminance Y is small (the neighborhood range containing a process pixel is dark). Conversely, the gain improvement function fg approaches 1 as the luminance Y increases (the neighborhood range including the processing pixel is brighter).

  The control unit 10 selects the gain enhancement function fg according to the brightness enhancement amount Gup described above, and increases the gain of the dark area by a method described later according to the brightness near the target pixel x, y.

  FIG. 5 illustrates gain improvement functions fg corresponding to lightness improvement amounts Gup = 0, 1, 2, and 3, respectively. Several gain improvement functions fg may be prepared in advance according to the lightness improvement amount Gup, or only a gain improvement function having the maximum gain increase amount is prepared, and the gain improvement function fg and gain 1 having the maximum gain increase amount are prepared. It is also possible to interpolate between the steps to make it stepless.

  Further, when it is determined in step 106 that the gradation compression mode is not set (the gradation non-compression mode is set), if the gradation compression process cannot be skipped due to the configuration of the processing circuit in the control unit 10, FIG. A gain improvement function fg shown in Gup = 0 is used.

  The gradation compression calculation in each pixel R [x, y], G [x, y], B [x, y] is performed by the following equations 4 to 6.

  Fg in Expression 4 to Expression 6 corresponds to the above-described gain improvement function fg, and Ly corresponds to the low-pass image described in Step 107.

  In step 111, the control unit 10 records the image data subjected to the gradation compression process in step 110 or the image data subjected to the normal image process in step 105 to the memory card 28 via the card interface 25, A series of processing ends. Note that before the image data is recorded on the memory card 28, an image compression process (such as a JPEG compression process) may be performed as necessary via the compression / decompression unit 26.

Next, the operation of the photometric calculation subroutine (step 101 in FIG. 3) will be described based on the flowchart in FIG.
First, in step 201, the control unit 10 drives the second imaging sensor 16 to acquire a pre-image of the object scene.

  In step 202, the exposure value calculation unit 10a of the control unit 10 calculates a temporary exposure control value BvCntl0 for shooting. As a specific calculation method, for example, the pixels of the second image sensor 16 are divided into 24 regions as shown in FIG. 7B, and an average output for each region is obtained.

A specific method for calculating the temporary exposure control value BvCntl0 will be described with reference to the flowchart shown in FIG.
First, in step 301, the control unit 10 acquires a 24-division photometric result from the second imaging sensor 16. The second image sensor 16 photoelectrically converts the incident light and, as shown in FIG. 9A, the 24 divided luminance values Bv [1,1] to Bv [6, corresponding to the divided areas. 4] is output.

  In step 302, the control unit 10 acquires a 15-division photometry result based on the 24-division photometry result acquired in step 301. The control unit 10 collects the four-divided luminance values Bv [1,1] to Bv [6,4] obtained in step 301 by four, and obtains the fifteen divided luminance values RBv [1] to RBv [15]. get. The 15-divided luminance values RBv [1] to RBv [15] are calculated by Equations 7 to 21 below. FIG. 9B illustrates a part of the 15 divided areas (RBv [1], RBv [3], RBv [5], RBv [11], RBv [13], RBv [15]).

RBv [1] = (Bv [1,1] + Bv [2,1] + Bv [1,2] + Bv [2,2]) / 4 (Expression 7)
RBv [2] = (Bv [2,1] + Bv [3,1] + Bv [2,2] + Bv [3,2]) / 4 ... (Equation 8)
RBv [3] = (Bv [3,1] + Bv [4,1] + Bv [3,2] + Bv [4,2]) / 4 ... (Equation 9)
RBv [4] = (Bv [4,1] + Bv [5,1] + Bv [4,2] + Bv [5,2]) / 4 ... (Equation 10)
RBv [5] = (Bv [5,1] + Bv [6,1] + Bv [5,2] + Bv [6,2]) / 4 ... (Equation 11)
RBv [6] = (Bv [1,2] + Bv [2,2] + Bv [1,3] + Bv [2,3]) / 4 ... (Equation 12)
RBv [7] = (Bv [2,2] + Bv [3,2] + Bv [2,3] + Bv [3,3]) / 4 ... (Equation 13)
RBv [8] = (Bv [3,2] + Bv [4,2] + Bv [3,3] + Bv [4,3]) / 4 ... (Formula 14)
RBv [9] = (Bv [4,2] + Bv [5,2] + Bv [4,3] + Bv [5,3]) / 4 ... (Equation 15)
RBv [10] = (Bv [5,2] + Bv [6,2] + Bv [5,3] + Bv [6,3]) / 4 ... (Equation 16)
RBv [11] = (Bv [1,3] + Bv [2,3] + Bv [1,4] + Bv [2,4]) / 4 (Expression 17)
RBv [12] = (Bv [2,3] + Bv [3,3] + Bv [2,4] + Bv [3,4]) / 4 ... (Equation 18)
RBv [13] = (Bv [3,3] + Bv [4,3] + Bv [3,4] + Bv [4,4]) / 4 ... (Equation 19)
RBv [14] = (Bv [4,3] + Bv [5,3] + Bv [4,4] + Bv [5,4]) / 4 ... (Equation 20)
RBv [15] = (Bv [5,3] + Bv [6,3] + Bv [5,4] + Bv [6,4]) / 4 (Expression 21)

  In step 303, the control unit 10 determines the average luminance value BvMean, the maximum luminance value BvMax15 of the luminance values RBv of 15 divisions, the minimum luminance value BvMin15, and the central luminance value BvC based on the photometric results obtained in steps 301 and 302. The maximum luminance value BvMax24 of the 24 divided luminance values Bv is calculated as the feature amount. Each value is obtained by the following equations.

BvMax15 = MAX (RBv [1] to RBv [15]) (Equation 23)
BvMin15 = MIN (RBv [1] to RBv [15]) (Equation 24)
BvC = RBv [8] (Equation 25)
BvMax24 = MAX (Bv [1,1] to Bv [6,4]) (Equation 26)

  In step 304, the control unit 10 calculates a temporary exposure control value BvCntl0 based on each value calculated in step 303. The temporary exposure control value BvCntl0 is obtained by the following equation.

BvCntl0 = k1, BvMean + k2, BvMax15 + k3, BvMin15 + k4, BvC + k5 (Equation 27)
In Equation 27, k1 to k4 are coefficients indicating the weight of each term. K5 is a constant term. k1 to k5 are numbers corresponding to the average luminance value BvMean, and are determined in advance so that a better image is obtained in various sample scenes. Examples of k1 to k5 are shown in Table 4 below.

Returning to FIG. 6, in Step 203, the restoration amount calculation unit 10 b of the control unit 10 calculates the temporary highlight restoration amount Pre_HiRcv. The calculation of the temporary highlight restoration amount Pre_HiRcv is performed according to steps 401 and 402 in FIG.
In step 401 in FIG. 10, the control unit 10 compares the temporary exposure control value BvCntl0 calculated in step 304 with the maximum luminance value BvMax24 of the 24 divided luminance values Bv. The control unit 10 obtains a value dHi in which the difference between the temporary exposure control value BvCntl0 and the maximum luminance value BvMax24 of the 24 divided luminance values Bv exceeds a predetermined value thHi using the following equation.

dHi = BvMax24-BvCntl0-thHi ... (Formula 28)
In Expression 28, the predetermined value thHi is an amount estimated to cause the highlight portion of the image to be saturated, and is, for example, about 2Ev to 3Ev. The predetermined value thHi is set to an optimum value depending on the saturation level of the first image sensor 9 which is an image sensor, the pixel size of the second image sensor 16, and the like.

  In step 402, the control unit 10 calculates a temporary highlight restoration amount Pre_HiRcv based on the comparison result in step 401. The control unit 10 obtains the temporary highlight restoration amount Pre_HiRcv using the following equation.

  As shown in Expression 29, the control unit 10 sets the value dHi obtained in step 401 from 0, where the difference between the temporary exposure control value BvCntl0 and the maximum luminance value BvMax24 of the 24 divided luminance values Bv exceeds the predetermined value thHi. Clip to the threshold value thdHi to set the temporary highlight restoration amount Pre_HiRcv.

  In Expression 29, the threshold thdHi is a predetermined threshold indicating the amount of highlight restoration that depends on the dynamic range of the first image sensor 9 that is the image sensor. The control unit 10 changes the threshold thdHi according to the shooting mode and the ISO sensitivity.

  The control unit 10 obtains a threshold thdHi corresponding to the shooting mode based on Table 5 and obtains a threshold thdHi corresponding to ISO sensitivity based on Table 6. The larger threshold thdHi is used to calculate the temporary highlight restoration amount Pre_HiRcv described above.

  When a person shooting mode is set in which it can be estimated that the main subject is a person, it is necessary to avoid unnecessarily dark shooting of the face portion of the person. Therefore, when the person photographing mode is set, as shown in Table 5, the value of the threshold thdHi is decreased.

  Further, when the first image sensor 9 is set to high ISO sensitivity, there is a possibility that noise increases and the dynamic range is insufficient. Therefore, when the first imaging sensor 9 is set to high ISO sensitivity, as shown in Table 6, the value of the threshold thdHi is decreased.

  Furthermore, a configuration may be adopted in which an expected amount of noise generation at the time of imaging is detected as appropriate, and an upper limit is set for the temporary highlight restoration amount Pre_HiRcv or the threshold thdHi described above is reduced according to the detection result.

  Returning to FIG. 6, in step 204, the main subject detection unit 10 c of the control unit 10 detects the main subject from the output of the second imaging sensor. For example, human face detection. Since the specific method is the same as that of a known technique, the description is omitted.

In step 205, the restoration amount restriction unit 10e of the control unit 10 calculates a highlight restoration amount restriction value Lmt_HiRcv.
Lmt_HiRcv = BvObj- (BvCntl0-BvSth) (Equation 30)

  Here, BvObj is the representative luminance value of the main subject detected in step 204. For example, when a scene as shown in FIG. 7C is photographed, the luminance is obtained from the pixel output at the center of the face as in the white area in FIG. 7D so as not to be affected by the background.

  The specific calculation is performed by the main subject luminance calculation unit 10d of the control unit 10 as follows.

(A) When the luminance of the main subject and the background are extremely different (for example, a scene of backlight or excessively forward light), an area wider than the face area (area FA0 in FIGS. 11A and 11B) When the representative luminance value BvObj is obtained from the above, there are cases where the appropriate face luminance cannot be obtained due to the influence of the background.

  Therefore, the control unit 10 may obtain the representative luminance value BvObj from an area inscribed in the outline of the person's face (area FA1 shown in FIG. 11A). Alternatively, the control unit 10 may obtain the representative luminance value BvObj from an area (area FA2 shown in FIG. 11B) whose size is smaller than the area circumscribing the outline of the person's face.

  Here, FIG. 11B shows an example in which the center of gravity of the area FA2 is arranged so as to overlap the center of gravity of the area FA0. However, the center of gravity of the area FA2 may be shifted below the center of gravity of the area FA0 in order to make it less susceptible to the influence of hair (illustration is omitted in this case). Further, FIG. 11B shows an example in which the size of the area FA2 is set to 1/2 the size of the area FA0. However, the size of the area FA2 can be appropriately set within a range not affected by the background.

(B) When the luminance difference in the face area is large (for example, a half-backlit scene in which light is applied to half of the face and half of the face is dark), the representative luminance is based on the average luminance value of the face area. When the value BvObj is obtained, the bright part of the face may be overexposed. Therefore, the control unit 10 may measure the face of the main subject by dividing the face into a plurality of areas, and obtain the representative luminance value BvObj based on a plurality of photometric values obtained for each of the plurality of areas.

As an example, the control unit 10 divides the face area into four so as to be 2 × 2 vertically and horizontally as shown in FIG. And the control part 10 is good also considering the maximum value among the photometry values of each division area (FA41-FA44) as a representative luminance value BvObj by the following formula | equation (31).
BvObj = Max (BvFA41, BvFA42, BvFA43, BvFA44) ... (Formula 31)
Here, in this specification, “BvBVFA41-44” indicates the average luminance value of the divided area (BVFA41-44) of the face area. In this specification, “Max (A, B)” means a function that returns the maximum value of A and B.

Returning to Equation 30, BvSth is the permissible under face amount.
BvSth may be a fixed value obtained in advance or a variable depending on the size of the face.
A calculation example in the case of variables will be described with reference to FIG.
BvSth = ObjUdLv2- (ObjUdLv2-ObjUdLv1) / (SthObjSz2-SthObjSz1) * (ObjSize-SthObjSz1) (Equation 32)
However, BvSth is limited between ObjUdLv1 and ObjUdLv2.
If BvSth <ObjUdLv1, BvSth = ObjUdLv1 (Equation 33)
If BvSth> ObjUdLv2, BvSth = ObjUdLv2 (Formula 34)

Here, ObjSize is the detected face size, SthObjSz1 and SthObjSz2 are face size thresholds, and ObjUdLv1 and ObjUdLv2 are face image output underlevel threshold values.
In step 206, the restoration amount restriction unit 10e of the control unit 10 restricts the restoration amount and calculates the highlight restoration amount HiRcv.
HiRcv = Pre_HiRcv ... (Formula 35)
However,
If HiRcv> Lmt_HiRcv, then HiRcv = Lmt_HiRcv (Equation 36)
In step 207, the control unit 10 obtains an exposure control value BvCntl1 from the following equation.
BvCntl1 = BvCntl0 + HiRcv (Formula 37)

According to the above-described embodiment, for example, when the face is dark due to backlight, when BvObj becomes small, Lmt_HiRcv becomes smaller from Expression 33, and HiRcv becomes easier to be limited by Lmt_HiRcv from Expressions 35 and 6. Equation 37 in the restricted case is replaced as follows:
BvCntl1 = BvObj + BvSth ... (Formula 38)

  That is, the exposure control value BvCntl1 cannot take a value larger than the right side of Expression 38. Accordingly, the output level of the face portion of the image data captured by the first image sensor based on the exposure control value BvCntl1 in step 103 does not become lower than the “level lower by BvSth than the reference exposure level”. For this reason, the face image level in step 104 is low and face detection cannot be performed, and noise is not conspicuous in the face portion of the image after the gradation compression processing in step 110.

  In a backlit person scene, if the face is large, the amount of highlight restoration is likely to be limited, so that the face of the person does not have to be dark. In addition, when the face is small, that is, when there are many backgrounds to be reflected, the amount of highlight restoration is difficult to be limited, so that saturation of the highlight portion of the background can be suppressed.

  In the above embodiment, the exposure control value is obtained based on the output of the second image sensor 16 different from the first image sensor 9, but in the case of the live view mode in which the image signal of the first image sensor 9 is a finder. Instead of the output of the second image sensor 16, the 9 outputs of the first image sensor may be used.

Further, ObjSize in the above embodiment may normalize the size of the detected main subject by the size of the shooting screen.
ObjSize = ObjSize ÷ PicSize (Formula 39)
Here, PicSize is the size of the shooting screen by the first image sensor. For example, the diagonal length of the shooting screen may be used, or the length of the short side may be used.

  Even if the ratio of the shooting screen by the first image sensor is switched to 4: 3 or 16: 9 by standardizing the size of the main subject, or an arbitrary area is cut out by digital zoom, the main subject occupying the shooting screen By making the under allowance of the main subject about the same with respect to the size of the image, it is possible to obtain an image with no sense of incongruity. In addition, the under-allowance amount of the main subject can be made the same among a plurality of imaging devices including the second imaging sensors having different numbers of divisions.

(Deformation)
The present invention is not limited to the above-described embodiment, and various modifications and changes as described below are possible, and these are also within the scope of the present invention. For example, in the above-described embodiment, gradation correction may be performed on the entire screen, which is partly gradation correction based on a low-pass image.
In addition, although embodiment and a deformation | transformation form can also be used in combination as appropriate, detailed description is abbreviate | omitted. Further, the present invention is not limited to the embodiment described above.

  1: camera body, 2: lens unit, 3: photographing lens, 9: first imaging sensor, 10: control unit, 10a: exposure value calculation unit, 10b: restoration amount calculation unit, 10c: main subject detection unit, 10d: Main subject luminance calculation unit, 10e: restoration amount restriction unit, 16: second imaging sensor, ::,:,:, Pre_HiRcv: provisional highlight restoration amount, HiRcv: highlight restoration amount, BvCntl0: provisional exposure restriction value, BvCntl1 : Exposure limit value, BvObj: Representative luminance value of main subject

Claims (3)

A metering unit for metering the subject field into a plurality of areas;
A first exposure calculation unit that calculates a first exposure value based on a photometric result by the photometric unit;
A correction amount calculation unit for calculating a correction amount for restoring the bright portion of the subject with respect to the first exposure value based on the photometric result;
A main subject detection unit for detecting a main subject based on the photometric results;
A main subject luminance calculation unit for calculating the luminance of the main subject;
A correction amount limiting unit that limits the correction amount based on the luminance of the main subject;
A second exposure calculation unit that calculates a second exposure value obtained by correcting the first exposure value based on the limited correction amount;
An imaging unit that images the subject based on the second exposure value and generates image data;
If the correction amount is greater than or equal to a limit value, the limited correction amount is the limit value,
The image pickup apparatus, wherein the limit value is a value obtained by subtracting a face under allowance from the first exposure value and subtracting the representative luminance of a main subject. The imaging device according to claim 1,
An imaging apparatus having an image processing unit that corrects gradation so that a dark part of the image data is brightened. A first exposure calculation unit that calculates a first exposure value based on a photometry result obtained by a photometry unit that divides the object field into a plurality of areas and performs photometry ;
A first calculation unit that calculates a correction amount for restoring a bright portion of the subject with respect to the first exposure value based on the photometric result ;
A second calculation unit for calculating a limit value for limiting the correction amount based on the luminance of the main subject ;
A selector for selecting said limit value when said limit value is smaller than the correction amount,
A second exposure calculation unit that calculates a second exposure value obtained by correcting the first exposure value based on the limit value selected by the selection unit when the limit value is smaller than the correction amount ;
An imaging unit that captures an image of a subject based on the second exposure value and generates image data;
An image processing unit that corrects gradation so that the image of the image data becomes brighter ,
The image pickup apparatus , wherein the limit value is a value obtained by subtracting a face under allowance from the first exposure value and subtracting the luminance of the main subject .

Images