JP2013040994A - Imaging apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an imaging apparatus capable of performing an appropriate AF even when a face exists the outside of a focus detection range.SOLUTION: An imaging apparatus 1 according to the present invention includes: face detection parts 25 and 42 that detect a face existing in an imaging area 50; an AF focus detection part 24 that includes a plurality of focus detection areas 51 in which a focal point can be detected in the imaging area; and a control part 40 that can decide that a focus detection area in which the focus detection is performed out of the focus detection areas in accordance with a face detection result in the face detection part. The control part classifies the focus detection areas depending on a column along which the focus detection area is arranged out of a plurality of columns that are parallel to one another and virtually provided in the imaging area, and even in a case in which a detected face does not exist at a position corresponding to the focus detection area, when the face is located at any column of the plurality of columns, a first control is performed that controls the AF focus detection part so that the focus detection is performed in the focus detection area decided by weighting a distance from the position in which the face is detected out of the focus detection areas belonging to the columns in which the face is located.

Description

  The present invention relates to an imaging apparatus.

  2. Description of the Related Art Conventionally, in a single-lens reflex type imaging apparatus, when a face such as a person is present in a focus detection area, focus detection is performed by a phase difference method in an area corresponding to the face (for example, see Patent Document 1).

JP 2008-61157 A

  However, the conventional imaging apparatus does not consider the case where the face is outside the range where the focus detection area is arranged.

  An object of the present invention is to provide an imaging apparatus capable of appropriate autofocus even when the face is outside the range where the focus detection area is arranged.

  The present invention solves the above problems by the following means. In addition, in order to make an understanding easy, although the code | symbol corresponding to embodiment of this invention is attached | subjected and demonstrated, it is not limited to this.

The invention described in claim 1 includes a face detection unit (25, 42) for detecting a face existing in the imaging region (50), and a focus detection region (51) capable of focus detection in the imaging region (50). It is possible to determine in which focus detection area of the focus detection area (51) the focus detection is performed according to the face detection results in the autofocus focus detection section (24) and the face detection sections (25, 42). A control unit (40), wherein the control unit (40) includes the focus detection area (51) in a plurality of parallel columns virtually provided in the imaging area (50). Even if the detected face does not exist at a position corresponding to the focus detection area (51), it is located in any one of the columns. If the face is located before Among the focus detection areas (51) belonging to the column, the autofocus focus detection section (24) performs focus detection in a focus detection area determined by giving a weight to the distance from the position where the face is detected. The imaging device (1) is characterized by performing first control.
Invention of Claim 2 is an imaging device (1) of Claim 1, Comprising: The said row | line | column also includes the row | line | column in which the said focus detection area | region (51) is not arrange | positioned, The said control part (40) When the face is located only in a row where the focus detection area (51) is not arranged, from the position where the face is detected in the focus detection area (51) belonging to the row close to the face. The imaging apparatus (1) is characterized in that the first control is performed so that focus detection is performed in a focus detection region (51) determined by giving a weight to the distance of the first focus.
Invention of Claim 3 is the imaging device (1) of Claim 1 or 2, Comprising: The detection result of the face by the said face detection part (25,42) contains the information regarding reliability, The said The control unit (40) is the imaging device (1) characterized in that the first control is not performed when the reliability of the detected face is smaller than a predetermined threshold.
Invention of Claim 4 is an imaging device (1) of any one of Claim 1 to 3, Comprising: The detection result of the face by the said face detection part (25,42) is related to reliability. An imaging apparatus (1) characterized in that when a plurality of faces including information are detected, the first control is performed based on the face with the highest reliability.
Invention of Claim 5 is an imaging device (1) of any one of Claim 1 to 4, Comprising: The said face detection part (25,42) contains a photometry sensor (25). The imaging device (1) characterized by the above.
Note that the configuration described with reference numerals may be modified as appropriate, and at least a part of the configuration may be replaced with another component.

  According to the present invention, it is possible to provide an imaging apparatus capable of appropriate autofocus even when a face is outside the focus detection range.

It is a figure explaining schematic structure of the camera of this embodiment. It is a block diagram which shows a control part and a photometry sensor. It is a figure which shows the focus detection area in an imaging area. It is a figure explaining a virtual focal field. It is a figure which shows the example which has the detected face outside a focus detection area | region. It is a figure explaining the principle of matching of a face detection result and a virtual focus detection area. It is a flowchart which shows the whole flow of this embodiment. It is a flowchart which shows the photometry process in the camera of the step of FIG. FIG. 9 is a flowchart for explaining a flowchart of auto area AF face detection calculation in the step of FIG. 8; FIG.

(Basic camera configuration)
Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a diagram illustrating a schematic configuration of a camera 1 according to the present embodiment.
The camera 1 according to the present embodiment is a so-called digital single-lens reflex camera including a camera body 20 and a photographing lens 10 that can be attached to and detached from the camera body 20.
The photographic lens 10 includes a plurality of lens groups constituting an imaging optical system and a diaphragm 11 inside the lens barrel. Of the plurality of lens groups, the illustrated lens group L1 is a focus adjustment lens that can move in the direction of the optical axis OA and adjust the imaging position.

  The camera body 20 includes an image sensor 21, a quick return mirror 22, a sub mirror 23, a finder optical system 30, a distance measuring element 24, a photometric sensor 25, and a control unit 40.

  The imaging element 21 is a photoelectric conversion element such as a CCD or CMOS that converts subject light into an electrical signal, and each pixel of red (R), green (G), and blue (B) is arranged in a predetermined arrangement pattern. ing. The imaging element 21 converts the image information on the imaging surface by the imaging optical system of the photographic lens 10 into an electrical signal and images it, and sends an image signal corresponding to the color information and luminance information corresponding to each pixel to the control unit 40. Output.

The quick return mirror (main mirror) 22 is provided so as to be movable between an action position interposed in the optical path from the imaging optical system of the photographing lens 10 to the image sensor 21 and a retracted position not interposed in the optical path. Yes. The quick return mirror 22 reflects the incident light beam to the finder optical system 30 (diffuse screen 26) at the operating position.
In addition, a semi-transmission region that transmits part of the incident light beam is formed in a part of the quick return mirror 22.
The sub mirror 23 is provided on the back side of the quick return mirror 22. The sub mirror 23 reflects the incident light beam transmitted through the semi-transmissive region of the quick return mirror 22 toward the distance measuring element 24.

The viewfinder optical system 30 includes a diffusing screen 26, a condenser lens 27, a pentaprism 28, and an eyepiece lens 29.
The diffusing screen 26 is provided at a position optically equivalent to the image sensor 21, and an incident light beam guided by the quick return mirror 22 forms an image.
The pentaprism 28 and the eyepiece lens 29 are formed so that the photographer can visually recognize the subject image formed on the diffusion screen 26 and passing through the condenser lens 27 as an erect image.

  Similarly to the image sensor 21, the photometric sensor 25 is a photoelectric conversion element such as a CCD or CMOS that converts subject light into an electrical signal, and each pixel of red (R), green (G), and blue (B) They are arranged in a predetermined arrangement pattern. The photometric sensor 25 has a structure in which the object scene is divided and photometric, and each photometric value can be output to the control unit 40.

The control unit 40 is a part that performs each control of the camera 1. FIG. 2 is a block diagram showing the control unit 40 and the photometric sensor 25. The control unit 40 includes a microcomputer 41 and an image processing IC (ASIC) 42.
The microcomputer 41 performs serial communication with the photometric sensor 25 and the ASIC 42 and controls them.
The ASIC 42 performs photometric calculation, scene recognition including face detection, and the like based on the photometric result from the photometric sensor 25.

Returning to FIG. 1, the distance measuring element 24 detects the imaging state of the subject image based on the subject light incident through the sub mirror 23, and outputs the detection information to the microcomputer 41 of the control unit 40.
As shown in FIG. 3, the distance measuring element 24 includes sensor units respectively arranged corresponding to a plurality of focus detection areas 51 set at predetermined positions in the imaging area 50. FIG. 3 shows an example when the focus detection area is 51 points.

(Basic camera operation)
Light passing through the taking lens 10 and passing through the diaphragm 11 is guided to the quick return mirror 22. One of the lights separated by the click return mirror 22 is guided to the sub mirror 23 and enters the distance measuring element 24.
The other light is projected onto the diffusing screen 26, passes through the condenser lens 27, is guided to the pentaprism 28, and proceeds to the eyepiece lens 29 that is a finder.
An image that has passed through the triangular prism 32 that is a photometric optical element by the pentaprism 28 is imaged on the photometric sensor 25 through the photometric lens 31.
The control unit 40 drives the photometric sensor 25 from the microcomputer 41 that performs sequence control of the camera 1 via serial communication.

The photometric sensor 25 performs a storing operation in accordance with a command from the microcomputer 41 and outputs photometric information A / D converted inside the photometric sensor 25 to the ASIC 42.
The ASIC 42 receives the output of the photometric sensor 25 and performs photometric and face detection processing.
Here, the face detection is performed using a known method, and the face information, for example, the position, size, inclination, and reliability of the face (for example, the degree of certainty of the face represented by 1 to 9). ) Etc. are detected.

  Then, when the camera 1 is in automatic selection AF (auto area AF), the ASIC 42 calculates a focus detection face area in which a face is detected, and transmits it to the microcomputer 41 via serial communication together with the face information.

  FIG. 3 is a diagram illustrating the focus detection area 51 in the imaging area 50. As shown in the figure, when the face is detected at the position indicated by the symbol A, the ASIC 42 transmits position information of the focus detection area 51a to the microcomputer 41.

  The microcomputer 41 controls the distance measuring element 24 to perform focus detection in the focus detection area 51a. The distance measuring element 24 determines the in-focus state using the phase difference by the sensor unit arranged in the focus detection face area 51a. The in-focus state is transmitted from the distance measuring element 24 to the microcomputer 41, and the microcomputer 41 drives the focusing lens L1 to focus.

When a release button (not shown) is pressed, the quick return mirror 22 jumps upward, and the subject light is guided to the image sensor 21 via the shutter 33, and a focused photographed image is obtained.
In live view, the quick turn mirror 22 is raised by operating a live view display switch (not shown) on the back, and is guided to the image sensor 21 to obtain a through image.

  As described above, the ASIC 42 performs face detection, and the microcomputer 41 transmits information on the focus detection area where the face is detected to the distance measuring element 24. The distance measuring element 24 measures the focus detection area at that position.

However, in this case, in the imaging region 50, there is a case where the face is present not at the portion where the focus detection region 51 as described above is arranged but at the position indicated by B in FIG. 3 outside the portion.
In such a case, since the focus detection area 51 corresponding to the face position does not exist, distance measurement by the distance measuring element 24 based on the face position cannot be performed.
Therefore, in this embodiment, a focus detection area for distance measurement is determined as follows.

(Determining the focus detection area for ranging when the face is outside the focus detection area)
First, coordinates obtained by extending the focus detection area in the vertical and horizontal directions are associated as virtual focus areas. FIG. 4 is a diagram for explaining the virtual focal region.
As shown in the drawing, the focus detection areas 51 are not evenly arranged on the entire surface of the imaging area 50, but rather are arranged at a slightly central portion. Therefore, there is a portion where the focus detection region 51 is not arranged in the portion from the outer edge of the imaging region.
Therefore, in this embodiment, a virtual focus area is set by extending the focus detection area vertically and horizontally.

As shown in FIG. 4A, sixteen lines having a predetermined width indicated by numbers 0 to 15 in the drawing along the direction in which the focus detection areas 51 are arranged in the vertical and horizontal directions of the imaging area. A virtual focus detection area shown in FIG.
Further, as shown in FIG. 4 (b), the left (outside) of the above-mentioned 0th line, the right (outside) of the 10th line, the top (outside) of the 11th line and the bottom of the 15th line ( Also on the outer side, virtual focus detection areas indicated by four lines having a predetermined width indicated by numbers 16 to 19 are provided. As a result, it is possible to associate with the four corners of the imaging area.

FIG. 5 is a diagram illustrating an example in which the detected face is outside the focus detection area.
For example, as shown in FIG. 5A, when the camera 1 is in the normal position (when taking a landscape image), the face is above the imaging area 50 outside the arrangement range of the focus detection area 51 (Y plus direction). May be located.
For example, in the case of FIG. 5A, the face exists on the 5th and 18th lines of the virtual focus detection area.
Therefore, the area outside face flag [5] = 1, the area outside face flag [18] = 1, and the others are set to 0.
Further, the reliability of the face detection result obtained at the time of face detection by the ASIC 42 is output to the out-of-area face reliability [5] and the out-of-area reliability [18] as reliability data.

For example, as shown in FIG. 5B, when the camera is in the vertical position (when shooting a vertically long image), the face is located above the imaging area outside the arrangement range of the focus detection area 51 (X plus direction). There is a case.
In this case, the face exists on the 12th, 13th, and 17th lines of the virtual focus detection area.
Therefore, the area outside face flag [12] = 1, the area outside face flag [13] = 1, the area outside face flag [17] = 1, and the others are set to 0. In addition, the reliability of the face detection result is output to the out-of-area face reliability [5], the out-of-area face reliability [13], and the out-of-area reliability [17] as reliability data.

  FIG. 5C shows a case where 18 to 16 lines are not applicable to the 0 to 10 lines of the virtual focus detection area. In this case, the body part may be in the focus detection area 51. The area outside face flag [18] = 1 and the area outside face flag [16] = 1 are set, and the face reliability is output to the outside area reliability [18] and the outside area reliability [16].

FIG. 5D shows a case where there are a plurality of faces outside the focus detection area.
For the area outside face flag, [2], [5], [8], [18] are set to 1, and the others are set to 0.
As for the out-of-area reliability, the reliability of each face is output for [2], [5], and [8], and the reliability that is the maximum among [2], [5], and [8] for [18]. Is output.

Next, determination of the focus detection area for distance measurement in FIGS. 5A to 5B will be described.
When the camera shown in FIG. 5A is in the horizontal position, as described above, the area outside face flag [5] = 1 and the area outside face flag [18] = 1. When the out-of-area reliability [18] obtained from the face detection macro is larger than a predetermined threshold, the focus detection operation is performed in the focus detection area in the Y minus direction of the out-of-area face flag [18].

  Here, the case where it is larger than the predetermined threshold is, for example, a case where the reliability is represented by a numerical value of 0 to 100, the threshold is 40, and the reliability of the out-of-area face flag [18] is 5.

  Further, the focus detection area in the Y minus direction of the out-of-area face flag [18] is located in the Y minus direction of the out-of-area face flag [18] in the five lines in which the face is detected in FIG. In addition, from the side closer to the 18th line, for example, the focus detection region having the highest contrast among the three focus detection regions 51a-1, 51a-2, and 51a-3.

  In addition to the case shown in FIG. 5 (a), in addition to the area outside face flag [18], [0] to [4] and [6] to [10] other than [5] are set to 1. The same applies to the case of.

  When the camera shown in FIG. 5B is in the vertical position, as described above, the area outside face flag [12] = 1, the area outside face flag [13] = 1, and the area outside face flag [17] = 1. . When the out-of-area reliability [17] obtained from the face detection macro is larger than a predetermined threshold value, the focus detection areas 51b-1, 51b-2, and 51b-3 in the X minus direction of the out-of-area face flag [17] The focus detection operation is performed in the focus detection region with the highest contrast.

In addition to the case shown in FIG. 5B, in addition to the area outside face flag of [18], [11], [14], and [15] other than [12] and [13] are 1 The same applies to the case of.
The same applies to the case where the out-of-area face flag is not [18] but [16].

  When the face corresponds to 18 lines and 16 lines as shown in FIG. 5C, and the out-of-area reliability [18] is larger than a predetermined threshold, the out-of-area face flag [18] of 0 line in the Y minus direction For example, three of the focus detection areas 51c-1, 51c-2, and 51c-3 having the highest contrast from the 18 lines (the focus detection areas of all 0 lines because there are three in the figure) are the highest in contrast. A focus detection operation is performed in the focus detection area.

  When there are a plurality of faces outside the focus detection area shown in FIG. 5D, 18 lines of the most reliable line (explained as 8 lines, for example) out of 2 lines, 5 lines, and 8 lines. For example, the focus detection operation is performed in the focus detection region having the highest contrast among the three focus detection regions 51d-1, 51d-2, and 51d-3.

(Principle of correspondence between face detection result and virtual focus detection area)
FIG. 6 is a diagram for explaining the principle of associating the face detection result with the virtual focus detection area.
The camera 1 adjusts the position of an imaging and focus detection position, a finder optical system, and the like with reference to the center of the optical axis of the mount. Therefore, as shown in FIGS. 4 and 5, the face detection area is replaced with the XY coordinates of the center of the optical axis, and the case where the center of the focus detection position enters the face area is defined as the focus detection face area.

When the center coordinates of the focus detection area are inside the circle whose radius is ½ of the face width from the face center coordinates, the focus area is set to the focus detection and the flag is set.
In the case shown in FIG. 6A, the right two focus detection areas 51e-1 and 51e-2 are set as the face focus detection areas, and the left two focus detection areas 51e-3 and 51e-4. Is not a focus detection area corresponding to the position of the face.

In the case of the virtual focus detection area when the camera 1 is at the normal position (FIG. 6B), the X coordinate of 0 to 10 lines enters a circle with a radius of 1/2 of the face width from the center coordinate of the face. Sometimes set the area face flag.
For the 18th line, a flag is set when the Y coordinate of the lower end of the circle whose radius is 1/2 of the face width from the center coordinate of the face is larger than the Y coordinate of the 11th line (Y plus direction).

FIG. 7 is a flowchart showing the overall flow of this embodiment.
First, when a release button (not shown) is half-pressed (step 11), photometry (step 12) and AF calculation (step 13) are performed.
During half-pressing, the processing from step 11 to step 13 is repeated (step 14, NO).
After a half-press release (step 11, NO), when a predetermined time has passed (step 21, YES), the power SW is turned off.
When the release button is pressed halfway down (step 14, YES), the mirror return operation of the quick return mirror 22 is performed (step 15), and the image sensor 21 is initialized (step 16). Real photographing is performed (step 17). In the actual shooting, the shutter curtain of the shutter 33 is held by the magnet, and the front curtain and the rear curtain travel for a predetermined time and are exposed to the image sensor 21.
When the actual photographing is finished, the quick return mirror 22 is lowered (step 18).
The photographed image is calculated and recorded by the image processing circuit of the ASIC 42 (step 19).

FIG. 8 is a flowchart showing the photometric process in the camera 1 in step 12 of FIG.
First, the photometric sensor 25 performs accumulation for performing photometry (step 21). In photometry accumulation, accumulation is performed so that the maximum luminance of the object scene becomes the target output level. This is because the luminance value in the range from the maximum luminance to the dynamic range is used.
Next, the photometric sensor 25 determines the validity of the accumulated data (step 22). In the case of data that does not deserve photometric calculation (step 23, NO), re-accumulation is performed as ineffective.
When there is validity (step 23, YES), photometry calculation is performed in the ASIC 41 (step 24).
Next, accumulation for face detection is performed in the photometric sensor 25 (step 25). This is an accumulation performed so that the brightness of the face becomes an appropriate level. A description of these accumulation controls is omitted.
Then, face detection is performed in the ASIC 41 (step 26). Face detection is based on the template matching method, and information such as face position, face size, reliability, etc. can be obtained.
Next, when the camera 1 is in automatic selection AF (auto area AF), auto area calculation is performed (step 27).

FIG. 9 is a flowchart for explaining the flowchart of the automatic area AF face detection calculation in step 27 of FIG. The auto area AF face detection calculation is a process of associating the face detection result with the focus detection face area 51.
First, the reliability of face detection is calculated from a time-series history, face position, face size, etc. by a known method (step 31).

Next, narrowing down is performed when there are a plurality of faces (step 32). This is because a more reliable face is used as the priority face area.
Next, the distance of the focus detection area 51 from the face center is calculated (step 33). At this time, an area where radius = face width / 2 × coefficient is set as a focus detection face area 51.
From the above, a face detection AF correspondence mask is created for the focus detection face region 51 (step 34). With the face detection AF-compatible mask, when the focus detection area is 51 points, the focus detection face area is set to 1 as 0 or 1 flag information for 51 bits, and the others are set to 0.

Next, when there is no face in the focus detection area 51 (step 35, YES), processing based on this embodiment is performed.
If there is a face other than the focus detection area 51, a flag is set when the face enters the virtual focus detection area as described with reference to FIG. 4, and is calculated as flag information (mask outside the face detection area) (step 36). .
Next, the reliability outside the focus detection range is calculated (step 37).
As described above, the focus adjustment is performed by driving the lens with reference to the reliability and the face detection AF compatible mask as needed when performing autofocus.

As described above, this embodiment has the following effects.
(1) In this embodiment, even if the face is not in the focus detection area and cannot be focused on, it is a portion where there is a possibility that a body suitable for focusing is present next to the face. Since focusing is performed, a photographed image close to the intention of the photographer who wants to focus on the face can be obtained.

(2) When the face does not exist in the focus detection area, the focus detection candidates are not one but three focus detection areas, and focus detection is performed in the focus detection area having the highest contrast. Therefore, since the contrast is low, the possibility that the focus cannot be detected is low.

(3) For example, even when a face exists in an area that is not on a virtual line including the focus detection area, such as the four corners of the imaging area, the focus detection area on the line close to the face Since focus detection is performed, it is possible to obtain a photographed image close to the photographer's intention intended to focus on the face.

(4) When the reliability is low in the face detection result, focus detection based on the face detection result is not performed, so that there is a high possibility that focus detection is performed on objects other than the face and body.
(5) When there are a plurality of faces in the face detection result, the face with the highest reliability is prioritized, so that more accurate focus detection can be performed.
(6) In this embodiment, when the virtual focus detection area is within a radius of ½ of the face size from the face center coordinate of the face detection result, flag information and reliability are transmitted to the microcomputer, so that the data amount is suppressed. It becomes possible.

(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 this embodiment, the focus detection operation is performed in the focus detection region having the highest contrast among the three focus detection regions close to the face. However, the present invention is not limited to this, and focus detection is performed in the closest focus detection region. May be. Further, the focus detection operation may be performed in the focus detection region having the highest contrast among the two or four or more focus detection regions.

Moreover, although the case where the focus detection area is 51 has been described in the present embodiment, the present invention is not limited thereto, and may be 50 or less and 52 or more. Further, the arrangement is not limited to the arrangement shown in FIGS.
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, 20: camera body, 24: distance measuring element, 25: photometric sensor, 40: control unit, 41: microcomputer, 42: ASIC, 50: imaging area, 51: focus detection area, L1: focusing lens

Claims (5)

A face detection unit for detecting a face existing in the imaging region;
An autofocus focus detection unit having a plurality of focus detection regions capable of focus detection in the imaging region;
A control unit capable of determining in which focus detection region of the focus detection region focus detection is performed according to a face detection result in the face detection unit;
With
The controller is
The focus detection areas are grouped according to which of a plurality of parallel columns virtually arranged in the imaging area and arranged, and the detected face is included in the focus detection area. Even if it does not exist in the corresponding position, if it is located in any one of the columns, the distance from the position where the face is detected in the focus detection region belonging to the column where the face is located Performing the first control to control the autofocus focus detection unit so as to perform focus detection in the focus detection region determined with a weight.
An imaging apparatus characterized by the above. The imaging apparatus according to claim 1,
The column also includes a column in which the focus detection area is not arranged,
The controller is
When the face is located only in a row where the focus detection area is not arranged, the distance from the position where the face is detected is weighted among the focus detection areas belonging to the row close to the face. Performing the first control so as to perform focus detection in the determined focus detection region;
An imaging apparatus characterized by the above. The imaging apparatus according to claim 1, wherein:
The face detection result by the face detection unit includes information on reliability,
The controller is
If the reliability of the detected face is less than a predetermined threshold, do not perform the first control;
An imaging apparatus characterized by the above. The imaging apparatus according to any one of claims 1 to 3,
The face detection result by the face detection unit includes information on reliability,
When a plurality of the faces are detected, performing the first control based on the face with the highest reliability;
An imaging apparatus characterized by the above. The imaging apparatus according to any one of claims 1 to 4, wherein:
The face detection unit includes a photometric sensor;
An imaging apparatus characterized by the above.

Images