JP2006227080A - Electronic camera - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an electronic camera by which a person being a subject is stably focused by face recognition. <P>SOLUTION: The electronic camera has: an imaging device to photoelectrically convert a subject image by a photographic optical system so as to generate the image signal of a photographic image plane; a face recognition part to detect a face area in the photographic image plane based on the image signal; a focus detection area designation part to set a focus detection area including the contour of the face area in a focus detection area group arranged within the photographic image plane as a designated area; and a focusing control part to calculate the focus evaluated value of the subject image based on the image signal corresponding to the designated area and detect the position of the photographic optical system where the focus evaluated value becomes maximum as a focusing position. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an electronic camera that performs focus detection based on a result of face recognition.

In recent years, electronic cameras that electronically record a subject image using an image sensor have been rapidly spreading, and such electronic cameras having a face recognition function are also known. For example, Patent Document 1 discloses an electronic camera that performs focusing control based on the eyes of a subject for an electronic camera having a face recognition function.
JP 2001-215403 A

However, in the electronic camera disclosed in Patent Document 1, it is difficult to focus on the subject's eyes when the subject person has closed eyes or wearing glasses, for example. There was still room for improvement in terms of low nature.
The present invention is to solve the above-described problems of the prior art, and an object of the present invention is to provide an electronic camera that can stably focus on a subject person through face recognition.

  An electronic camera according to a first aspect of the present invention is an image sensor that photoelectrically converts a subject image by a photographing optical system to generate an image signal of a photographing screen, and face recognition that detects a face area in the photographing screen based on the image signal. A focus detection area designating unit that sets, as a designated area, a focus detection area that includes the outline of the face region, among the focus detection area groups arranged in the shooting screen, and the image signal corresponding to the designated area And a focus control unit that calculates a focus evaluation value of the subject image based on the image and detects a position of the photographing optical system that maximizes the focus evaluation value as a focus position.

According to a second aspect, in the first aspect, the focus detection area designating unit sets a part of the plurality of focus detection areas including the outline of the face area as the designated area.
According to a third aspect, in the second aspect, the focus detection area designating unit sets a focus detection area overlapping with a contour on the upper side or the side of the face region as the designated area.

In a fourth aspect based on any one of the first to third aspects, the focus detection area designating unit is configured to detect the face when the focus position cannot be detected in a focus detection area including a contour of the face area. The focus detection area located below the area is changed to the designated area.
In a fifth aspect based on the third or fourth aspect, the face recognition unit detects a face direction based on a positional relationship of face parts in the face region, and the focus detection area designation unit includes the face The position of the focus detection area, which is the designated area, is changed according to the direction.

According to a sixth invention, in the third or fourth invention, the camera further includes a posture detection unit that detects a shooting posture of the electronic camera, and the focus detection area designating unit is configured to detect the designated area according to the photographing posture. The position of the focus detection area is changed.
According to a seventh invention, in any one of the first to sixth inventions, when a finder image of a shooting screen is displayed based on the image signal and the in-focus position cannot be detected in the designated area, the finder is displayed. It further has an electronic finder unit for displaying a non-focusable display associated with the face area of the image.

(Function)
In the first invention, a focus evaluation value of a subject image is calculated based on an image signal of a focus detection area (designated area) including a contour of a face region, and a so-called contrast detection type focusing operation is performed. That is, in the first invention, since the focus evaluation value is calculated using information on the contour portion of the face with high contrast, it is possible to improve the focusing accuracy of the detected subject on the face. Further, the detection of the contour of the face area is relatively easy, and the possibility that the focusing accuracy is affected by the facial expression of the subject is reduced.

In the second invention, a focusing operation is performed using an image signal of a part of the focus detection area including the outline of the face region. Therefore, in the second invention, in addition to the operation almost the same as that of the first invention, the calculation load of the focus evaluation value by the focusing control unit is reduced, so that the configuration of the focusing control unit can be simplified and the focusing operation can be performed. It is possible to further increase the speed.
In the third invention, when the designated area is set in the second invention, priority is given to the focus detection area overlapping with the upper contour or the lateral contour of the face region. That is, in the third invention, the lower part of the face area where the contrast is lowered due to the skin color part of the neck is excluded, and the place where the contrast is higher on the upper side or the side of the face area is set as the designated area. Also in the invention, high focusing accuracy can be ensured.

In the fourth invention, when the focus position cannot be detected in the focus detection area including the outline of the face area, the focus operation is performed in the focus detection area below the face area where the human body is located. Therefore, even if the focus position cannot be detected in the face area, the possibility that the subject person can be focused is improved.
In the fifth invention, the position of the designated area is changed by detecting the direction of the face based on the positional relationship of the facial parts in the face area. Therefore, stable focusing accuracy can be ensured regardless of changes in the photographing posture such as the normal position and the vertical position.

In the sixth invention, the position of the designated area is changed by the posture detection unit that detects the photographing posture of the electronic camera. Therefore, stable focusing accuracy can be ensured regardless of changes in the photographing posture such as the normal position and the vertical position.
In the seventh invention, when the in-focus position cannot be detected in the designated area, the in-focus indication associated with the face area of the finder image is displayed on the electronic finder unit. Therefore, the user can easily determine whether or not the subject person is in focus based on the display of the electronic viewfinder.

  According to the present invention, since the focus evaluation value is calculated using information on the face contour portion having high contrast and easy detection, it is possible to improve the focusing accuracy of the detected subject on the face.

(Description of the first embodiment)
FIG. 1 is a block diagram showing an outline of the electronic camera of the first embodiment. The electronic camera according to the first embodiment includes a photographing lens 11, a lens driving mechanism 12, an image sensor 13, an analog signal processing unit 14, an A / D conversion unit 15, an image processing unit 16, and a compression / decompression processing unit. 17, a memory 18, a card I / F 19, a monitor I / F 20 and a liquid crystal monitor 21, an operation unit 22, a CPU 23, and a data bus 24. The image processing unit 16, the compression / decompression processing unit 17, the memory 18, the card I / F 19, the monitor I / F 20, and the CPU 23 are connected via a data bus 24.

The taking lens 11 is composed of a plurality of lens groups including a focusing lens for adjusting the in-focus position. The photographing lens 11 is adjusted in position in the optical axis direction by a lens driving mechanism 12.
The image sensor 13 is disposed on the image space side of the photographing lens 11. On the light receiving surface of the image sensor 13 (the surface opposite to the photographing lens 11), light receiving pixels for photoelectrically converting a subject image to generate an analog image signal are two-dimensionally arranged. The output of the image sensor 13 is connected to an analog signal processing unit 14.

  In addition, the imaging device 13 exposes the subject at predetermined intervals even when the release is not performed, and outputs an analog image signal (through image signal) by thinning readout or the like. This through image signal is used for AF calculation, AE calculation and face recognition by the CPU 23, and generation of a finder moving image by the image processing unit 26. Note that the image sensor 13 of the first embodiment may be either a charge sequential transfer method (CCD or the like) or an XY address method (CMOS or the like).

  The analog signal processing unit 14 includes a CDS circuit that performs correlated double sampling, a gain circuit that amplifies the output of the analog image signal, a clamp circuit that clamps the waveform of the input signal to a constant voltage level, and the like. The A / D conversion unit 15 converts the analog image signal output from the analog signal processing unit 14 into a digital image signal. The output of the A / D conversion unit 15 is connected to the image processing unit 16 and the CPU 23, respectively.

The image processing unit 16 performs image processing (defective pixel correction, gamma correction, interpolation, color conversion, edge enhancement, etc.) on the digital image signal at the time of release to generate photographed image data. The image processing unit 16 sequentially generates a finder image based on the digital image signal (through image signal) at the time of non-release.
Further, the image processing unit 16 synthesizes and displays a rectangular AF frame indicating the AF target face area on the finder image based on face recognition information described later (see FIG. 4). Further, the image processing unit 16 performs an in-focus display on the finder image using the AF frame based on the in-focus inability information described later. Examples of the in-focus incapacity display include blinking display of the AF frame and means for changing the color of the AF frame during normal operation. As will be described later, when the AF calculation is executed twice and none of them is in focus, the image processing unit 16 performs different in-focus indications for the first time and the second time.

  The compression / decompression processing unit 17 executes processing for compressing the captured image data after image processing in the JPEG format, and processing for expanding and restoring the compressed image data in the JPEG format. The memory 18 is composed of SDRAM or the like, and has a capacity capable of recording image data for a plurality of frames. The memory 18 temporarily stores image data before and after image processing by the image processing unit 16.

  A connector for connecting the recording medium 25 is formed on the card I / F 19. The recording medium 25 is composed of a known semiconductor memory or the like, and the photographic image data is finally recorded on the recording medium 25. The photographed image data generated in the first embodiment conforms to the Exif (Exchangeable image file format for digital still cameras) standard, and the photographed image data body is associated with attached information (shooting information, etc.) on the photographed image data. Has been recorded.

A liquid crystal monitor 21 is connected to the monitor I / F 20. The liquid crystal monitor 21 is mainly disposed on the back portion of the electronic camera. The finder images sequentially output from the image processing unit 16 are displayed as moving images on the liquid crystal monitor 21 during shooting. The liquid crystal monitor 21 also displays a captured image data playback screen, a setting screen for changing various settings of the electronic camera, and the like.
The operation unit 22 includes an input button for performing switching and setting input of various shooting modes (such as a shooting mode and a playback mode) of the electronic camera, a release button, and the like.

  The CPU 23 controls the operation of each part of the electronic camera according to a sequence program stored in a ROM (not shown). For example, the CPU 23 executes AE calculation, white balance gain calculation, and the like based on the through image signal. Further, the CPU 23 generates attached information of the captured image data based on the Exif standard at the time of release. And especially in 1st Embodiment, CPU23 has the following functions.

First, the CPU 23 performs a known face recognition process on the through image signal, and detects a human face area in the shooting screen. Then, the CPU 23 generates face recognition information indicating the position of the face area in the shooting screen. In the first embodiment, the CPU 23 also detects the vertical direction of the face based on the positional relationship of face parts (eyes, nose, mouth, etc.) during face recognition.
As an example of face recognition processing, Japanese Patent Application Laid-Open No. 8-63597 discloses a method for detecting a face with a degree of matching with a face outline template prepared in advance by extracting the outline of a skin color area based on color, In addition, a method for detecting the face using the matching degree with the eye template by obtaining the eye candidate area, the two-dimensional Fourier transform result of the face candidate area obtained from the face contour template, and the eyes, nose, mouth, etc. are prepared in advance. A method is disclosed in which a feature amount defined from the two-dimensional Fourier transform result of the face template image obtained is obtained, and the feature amount is subjected to threshold processing to perform face detection.

  Secondly, the CPU 23 executes a contrast detection AF operation based on a through image signal of a designated area located in the shooting screen. Here, the designated area is selected by the CPU 23 based on the face recognition information from among a plurality of focus detection areas (focus detection area group) regularly arranged in the shooting screen. In the first embodiment, all focus detection areas located in a rectangular area surrounding the outline of the face area constitute a designated area. In the first embodiment, the designated area is set to coincide with the range of the AF frame. Since it is rare that the range of the designated area and the face area completely coincide with each other, the designated area includes a peripheral portion adjacent to the face area, and high contrast occurs in the outline portion of the face area. (See FIG. 3).

In the first embodiment, when the in-focus position cannot be detected by the first AF calculation (designated area including the face area), the CPU 23 detects the focus detection area (the body of the subject is located below the face area). The focus detection area that is likely to exist is changed to the designated area. The change setting of the designated area is based on the vertical direction of the face detected by the CPU 23 with the face parts.
Here, the AF calculation of the contrast detection method is performed based on the principle that “the degree of image blur and the contrast have a correlation, and the contrast of the image is maximized at the time of focusing”. Specifically, the CPU 23 first extracts a high-frequency component in a predetermined band from the through image signal corresponding to the designated area using a band-pass filter. Then, the CPU 23 integrates the absolute value of the high frequency component to generate a focus evaluation value related to the subject image in the designated area. This focus evaluation value becomes the maximum value when the contrast becomes maximum at the in-focus position.

  Next, the CPU 23 moves the focusing lens in a predetermined direction and compares the focus evaluation values before and after the movement. If the focus evaluation value after the movement is large, it is considered that the contrast tends to increase, and the CPU 23 further performs the same calculation by moving the focusing lens further in the same direction. On the other hand, when the focus evaluation value after the movement is small, the contrast is in a lowered state. Therefore, the CPU 23 performs the same calculation by moving the focusing lens in the reverse direction. By repeating the above process, the CPU 23 searches for the peak (focus position) of the focus evaluation value. The above operation is generally referred to as “mountain climbing operation”. If the in-focus position cannot be detected in the designated area, the CPU 23 outputs in-focus inability information to the image processing unit 16.

Hereinafter, the photographing operation in the first embodiment will be described with reference to the flowchart of FIG.
Step S101: The CPU 23 causes the image sensor 13 to generate a through image signal at predetermined intervals. The image processing unit 16 generates a finder image based on the through image signal, and the CPU 23 displays the finder image on the liquid crystal monitor 21 as a moving image. Therefore, the user can perform framing of the subject using the finder image displayed on the liquid crystal monitor 21.

Step S102: The CPU 23 determines whether or not the release button is half-pressed. When the release button is half-pressed (YES side), the process proceeds to S103. On the other hand, when there is no input to the release button (NO side), the CPU 23 waits for half pressing of the release button.
Step S103: The CPU 23 detects the face area of the subject in the shooting screen based on the through image signal, and generates face recognition information if there is a face area in the shooting screen.

Step S104: The CPU 23 determines whether or not a face area is detected in S103. If a face area is detected (YES side), the process proceeds to S105. On the other hand, when the face area is not detected (NO side), the process proceeds to S108.
Step S105: Based on the face recognition information (S103), the CPU 23 sets a focus detection area in a rectangular area surrounding the outline of the face area as a designated area (see FIG. 3). Then, the CPU 23 performs an AF calculation by a hill-climbing operation based on the through image signal in the designated area. At the time of AF calculation in S105, the image processing unit 16 compositely displays an AF frame on the face area in the finder image (see FIG. 4).

Step S106: The CPU 23 determines whether or not the in-focus position has been detected in the designated area (S105). If the in-focus position has been detected (YES side), the process proceeds to S109. On the other hand, if the in-focus position cannot be detected (NO side), the CPU 23 generates in-focus inability information and proceeds to S107.
Step S107: In this case, the CPU 23 changes the focus detection area located below the face area to the designated area. Then, the CPU 23 executes the AF calculation again in the designated area after the change, and then proceeds to S109. Even when the in-focus position cannot be detected by the second AF calculation, the CPU 23 generates in-focus inability information. Note that during the AF calculation in S107, the image processing unit 16 performs in-focus display using the AF frame of the finder image based on the first or second in-focus inability information.

Step S108: On the other hand, in this case, there is no person in the shooting screen, or the face of the subject person cannot be detected, so the CPU 23 selects the focus detection area by a normal operation and executes the AF calculation. .
Step S109: Then, when the user fully presses the release button, the CPU 23 captures the subject and generates captured image data. At the time of generating the photographed image data, the CPU 23 records the attached information such as “presence / absence of face recognition” and “position of the designated area used for the AF calculation” in the photographed image data by using an Exif MakerNote tag. .

  Step S110: The CPU 23 determines whether or not there is a photographing end instruction by a user input. When there is a photographing end instruction (YES side), the CPU 23 stops the generation of the through image signal and ends the photographing. On the other hand, when there is no instruction to end photographing (NO side), the process returns to S102, and the CPU 23 repeats a series of operations. This is the end of the description of the shooting operation of the first embodiment.

Next, the effect of the first embodiment will be described.
(1) In the first embodiment, the CPU 23 performs the AF calculation in the designated area including the outline of the face area, so that it is possible to easily focus on a person in the shooting screen. In particular, in the designated area, high contrast occurs in the contour portion of the face area. Therefore, compared with the case where focus detection is performed only in a portion having a low contrast in the face region, the first embodiment makes it easy to search for a peak of contrast, and thus the focusing accuracy on the face of the subject is improved. Further, since the detection of the contour of the face area is relatively easy, the possibility that the focusing accuracy is affected by the facial expression of the subject is reduced.

  (2) In the first embodiment, when the in-focus position cannot be detected by the first AF calculation using the face area as the designated area, the CPU 23 performs the second AF calculation in the designated area where the human body is located. (S107). Therefore, even when the subject cannot be focused by the face area, the subject person can be focused with high probability. Further, the CPU 23 estimates the position of the body from the face direction and sets the second designated area. Therefore, in the second AF calculation, stable focusing accuracy can be ensured regardless of the photographing posture such as the normal position and the vertical position of the electronic camera.

  (3) In the first embodiment, at the time of AF calculation, an AF frame is synthesized and displayed on the face area in the finder image (S105). Therefore, the user can easily grasp the person who is the AF target from the finder image on the liquid crystal monitor 21. Further, during the second AF calculation, an in-focus display using the AF frame is performed on the finder image (S107). Since the in-focus state display differs between the first in-focus state and the second in-focus state, it is relatively easy for the user to determine whether the person is in focus based on the AF frame display state. it can.

(4) In the first embodiment, “presence / absence of face recognition”, “position of designated area used for AF calculation”, and the like are recorded in the captured image data as attached information. Therefore, by referring to the attached information of the photographed image data by a viewer such as a personal computer, the user can grasp the situation at the time of photographing afterwards.
(Description of Second Embodiment)
FIG. 5 is a block diagram showing an outline of the electronic camera of the second embodiment. In the following description of the embodiments, the same components as those in the first embodiment are denoted by the same reference numerals, and description thereof is omitted.

  The second embodiment is a modification of the first embodiment, and differs in configuration from the first embodiment in that the attitude sensor 26 is connected to the CPU 23. The posture sensor 26 has a shooting posture with the electronic camera held at the normal position, a right upper vertical shooting posture with the right hand side of the electronic camera up, a left upper vertical shooting posture with the left hand side of the electronic camera up, and an electronic camera. An inverted reverse position shooting posture is detected. If the in-focus position cannot be detected in the designated area including the face area, the CPU 23 changes the focus detection area located below the face area to the designated area based on the output of the posture sensor 26.

In the second embodiment, substantially the same effect as in the first embodiment can be obtained. In addition, since the position of the designated area is changed by the output of the posture sensor 26, it is possible to reduce the calculation load of the CPU 23 related to the detection of the vertical direction of the face.
(Description of the third embodiment)
FIG. 6 is a flowchart showing the photographing operation in the third embodiment. Here, since each step of the third embodiment corresponds to each step of the first embodiment except for S205, a duplicate description is omitted. The block diagram of the electronic camera of the third embodiment is common to that of the first embodiment or the second embodiment, and is not shown.

Step S205: Based on the face recognition information (S203), the CPU 23 sets only a part of the focus detection areas among the focus detection areas overlapping the contour of the face area as the designated area (see FIG. 7). Then, the CPU 23 performs an AF calculation by a hill-climbing operation based on the through image signal in the designated area.
Even in S205, since the contour portion of the face area is set as the designated area, the focusing accuracy on the face of the subject can be improved as in the first embodiment. However, in the focus detection area that overlaps the lower contour of the face area, there is a high possibility that the contrast will be lower due to the skin color part of the neck. Therefore, in S205, it is preferable that the CPU 23 sets the focus detection area overlapping with the upper contour or the lateral contour of the face area as the designated area. When the focus detection area on the upper side or the lateral side of the face area is used, the designated area is selected based on the vertical direction of the face detected by the CPU 23 and the output of the posture sensor 26.

In the third embodiment, since the designated area is smaller than that in the first embodiment, the calculation amount during AF calculation is also reduced. Therefore, in the third embodiment, it is possible to simplify the circuit configuration of the CPU 23 and further speed up the AF calculation.
(Correspondence between Claims and Embodiments)
Here, the correspondence between the claims and the embodiment is shown. Note that the correspondence shown below is an interpretation given for reference only, and does not limit the technical scope of the present invention.

The first embodiment corresponds to claims 1, 4, 5, and 7. The second embodiment corresponds to the first, fourth, and sixth aspects. The third embodiment corresponds to claims 1, 2, 3, 5, and 6.
The photographing lens 11 corresponds to the “photographing optical system”. The imaging element 13 corresponds to “imaging element”. Each function of the CPU 23 corresponds to the “face recognition unit”, “focus detection area designation unit”, and “focus control unit”. The posture sensor 26 corresponds to the “posture detection unit”. The “electronic finder unit” mainly corresponds to the image processing unit 16, the monitor I / F 20, and the liquid crystal monitor 21.

(Supplementary items of the embodiment)
As mentioned above, although this invention has been demonstrated by said embodiment, the technical scope of this invention is not limited to the said embodiment.
In the above embodiment, when it is impossible to focus in the designated area including the face area, the second AF calculation may be performed regardless of the result of the face recognition in the focus detection area at the center of the shooting screen, for example. Further, in the third embodiment, when in-focus is impossible in the first designated area, the second AF calculation may be performed in another focus detection area of the face area. Furthermore, in the above-described embodiment, the in-focus state display may be performed only when the in-focus state is not possible in the second time.

1 is a block diagram showing an outline of an electronic camera according to a first embodiment. Flow chart showing photographing operation in the first embodiment The figure which shows the position of the designation | designated area in 1st Embodiment The figure which shows the finder image at the time of face recognition in 1st Embodiment The block diagram which shows the outline | summary of the electronic camera of 2nd Embodiment Flow chart showing photographing operation in the third embodiment The figure which shows the position of the designation | designated area of 3rd Embodiment

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Shooting lens 12 Lens drive mechanism 13 Image sensor 14 Analog signal processing part 15 A / D conversion part 16 Image processing part 17 Compression / decompression processing part 18 Memory 19 Card I / F
20 Monitor I / F
21 LCD monitor 22 Operation unit 23 CPU
24 Data bus 25 Recording medium 26 Attitude sensor

Claims (7)

An image sensor that photoelectrically converts a subject image by the photographing optical system to generate an image signal of the photographing screen;
A face recognition unit for detecting a face area in the shooting screen based on the image signal;
A focus detection area designating unit that sets, as a designated area, a focus detection area that includes the outline of the face region, among the focus detection area groups arranged in the shooting screen;
A focus control unit that calculates a focus evaluation value of the subject image based on the image signal corresponding to the designated area, and detects a position of the photographing optical system that maximizes the focus evaluation value;
An electronic camera comprising:   The electronic camera according to claim 1, wherein the focus detection area designating unit sets a part of a plurality of focus detection areas including an outline of the face area as the designated area.   The electronic camera according to claim 2, wherein the focus detection area designating unit sets a focus detection area overlapping with a contour on the upper side or the side of the face area as the designated area.   The focus detection area designating unit changes a focus detection area located below the face area to the designated area when the focus position cannot be detected in the focus detection area including the outline of the face area. The electronic camera according to any one of claims 1 to 3, wherein the electronic camera is characterized by the following. The face recognition unit detects a face direction based on a positional relationship of face parts in the face region;
5. The electronic camera according to claim 3, wherein the focus detection area designating unit changes a position of a focus detection area serving as the designated area according to a direction of the face. A posture detection unit for detecting a shooting posture of the electronic camera;
5. The electronic camera according to claim 3, wherein the focus detection area designating unit changes a position of a focus detection area serving as the designated area according to the photographing posture. An electronic finder unit that displays a finder image on the shooting screen based on the image signal and displays an in-focus display associated with the face area of the finder image when the in-focus position cannot be detected in the designated area. The electronic camera according to claim 1, further comprising:

Images