JP2008187677A - Digital camera - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To prevent a poster or the like from being recognized as a person in the case where a subject including the poster or the like, other than a person, is photographed by a digital camera. <P>SOLUTION: The digital camera comprises: an imaging device including a photoelectric transducer; an image data generating means for generating image data from an output signal of the imaging device; a face image extracting means for extracting a face image including a face from the image data generated by the image data generating means; a feature point extracting means for extracting feature points of the face included in the face image from the face image extracted by the face image extracting means; a feature point change detecting means for detecting a change with the passage of time in the feature points of the face extracted by the feature point extracting means; and a person determining means for determining whether the face included in the face image extracted by the face image extracting means is that of a person or not, based on the change with the passage of time in the feature points of the face detected by the feature point change detecting means. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a digital camera, and more particularly to a digital camera having a face recognition function.

  Conventionally, in a digital camera, a process of extracting an image including a face (also referred to as a face image) from a captured image has been performed (see, for example, Patent Document 1). In the digital camera described in Patent Document 1, color data for face recognition is stored in advance, the color data of the entire area of the photographed image is compared with the color data for face recognition, and a matching portion is detected by the person. Recognize as a face and cut out an area including the face of this person.

  The cut face image is used for various purposes. For example, it has been proposed to calculate the degree of coincidence between a face image including a face and data in the shape database unit, and extract feature points based on the degree of coincidence and shape data (see Patent Document 2). The extracted feature points are used for various purposes. For example, it has been proposed to detect the state of the driver by extracting feature points of the driver's eyes (see Patent Document 3).

In addition, Patent Documents 4 and 5 and the like are considered to be related to the present application.
JP-A-10-334213 JP 05-197793 A JP 2005-18651 A JP 05-68715 A JP 05-344504 A

  However, in the digital camera described in Patent Document 1 or the like, even when a poster other than a person is photographed, the poster is recognized as a person's face, and an area including the poster is cut out. There is a problem that cannot be performed properly.

  The present invention has been made in view of such circumstances, and an object of the present invention is to prevent a poster or the like from being recognized as a person when a subject including a poster or the like other than a person is photographed with a digital camera. And

  The present invention has been made to solve the above problems, and the invention according to claim 1 is an image sensor including a photoelectric conversion element, and image data generating means for generating image data from an output signal of the image sensor. A face image extracting means for extracting a face image including a face from the image data generated by the image data generating means, and a feature point of the face included in the face image from the face image extracted by the face image extracting means A feature point extracting unit for extracting the feature point, a feature point change detecting unit for detecting a temporal change in the feature point of the face extracted by the feature point extracting unit, and a face feature detected by the feature point change detecting unit Person determination means for determining whether or not the face included in the face image extracted by the face image extraction means is a person based on a change in point over time.

  According to the first aspect of the present invention, the facial feature points included in the facial image are extracted, and the temporal change in the facial feature points is detected.

  If the face included in the face image is a person, the extracted feature points of the face change over time (for example, if the feature points are eyes (the same applies to the mouth, etc.) Changes over time). On the other hand, if the face included in the face image is not a person (for example, if it is a poster, painting, figure), the extracted facial feature points do not change over time.

  Therefore, it is possible to determine whether or not the face included in the face image is a person based on the temporal change of the extracted facial feature points. That is, when a subject including a poster other than a person is photographed with a digital camera, the poster or the like can be prevented from being recognized as a person.

  According to a second aspect of the present invention, in the first aspect of the invention, the feature point extracting unit calculates a feature point of a face outline included in the face image from the face image extracted by the face image extracting unit. And the feature point change detecting means extracts the feature point of the face extracted by the feature point extracting means based on the change of the feature point with time of the contour of the face extracted by the feature point extracting means. And the person determining means is included in the face image extracted by the face image extracting means based on the temporal change of the facial feature points corrected by the feature point change detecting means. It is characterized by determining whether a face is a person.

  According to the second aspect of the present invention, even if the facial feature point extracted by camera shake changes, the amount of camera shake is corrected, and based on the temporal change of the corrected facial feature point, the facial image It is determined whether the face included in is a person.

  Therefore, erroneous determination due to camera shake can be reduced.

  The invention according to claim 3 is the invention according to claim 1, further comprising area detection means for detecting the area of the face image extracted by the face image extraction means, a feature point change detection means, and person determination The means is characterized in that each processing is executed when the area detected by the area detection means exceeds a certain value.

  According to the third aspect of the present invention, the process for determining whether or not the face included in the face image is a person is performed only when the area of the extracted face image exceeds a certain value.

  Therefore, it is possible to reduce misjudgment or useless processing due to the area of the face image not exceeding a certain value.

  According to a fourth aspect of the present invention, in the first aspect of the invention, the feature point is further provided with a focus unit that preferentially focuses a face included in the face image extracted by the face image extraction unit. The change detection unit and the determination unit execute the processes when the focus position by the focus unit is located at a predetermined position on the wide-angle side.

  According to the fourth aspect of the present invention, whether the face included in the face image is a person, such as extracting a facial feature point from the face image only when the focus position is located at a predetermined position on the wide-angle side. Processing for determining whether or not is performed.

  As a result, it is possible to target an image with less focus shift, and thus it is possible to reduce erroneous determination. Further, the processing can be reduced in weight.

  According to a fifth aspect of the present invention, in the first aspect of the invention, the storage means for storing the face feature points extracted by the feature point extracting means and the results determined by the person determining means in association with each other And an in-memory determination unit that determines whether or not the feature point of the face newly extracted by the feature point extraction unit is stored in the storage unit, and the feature point extraction unit newly determines by the in-memory determination unit When it is determined that the feature points of the face extracted in step S3 are stored in the storage unit, the determination result stored in association with the feature points of the face is read from the storage unit. To do.

  According to the fifth aspect of the present invention, when it is determined that the newly extracted facial feature point is stored in the storage unit, the facial feature point is stored in association with the facial feature point from the storage unit. The judgment result is read.

  That is, when it is determined that the newly extracted facial feature points are stored in the storage unit, it is not necessary to detect a change with time of the extracted facial feature points, so that it is extremely quick. In addition, it is possible to determine whether or not the person is a person.

  The invention according to claim 6 is the invention according to claim 1, further comprising display means for displaying a display representing a determination result by the person determination means on a display surface.

  According to the sixth aspect of the present invention, for example, in a so-called through image, a subject determined to be a person is surrounded by a dotted rectangular frame (for example, green) and displayed, and a subject determined not to be a person is displayed as a dotted rectangle. It can be displayed by being surrounded by a frame (for example, red). Alternatively, only a subject determined to be a person can be displayed surrounded by a rectangular frame.

  This makes it possible to instantaneously determine whether or not the subject on the display surface is a person.

  According to the present invention, when a subject including a poster or the like other than a person is photographed with a digital camera, the poster or the like can be prevented from being recognized as a person.

[First embodiment]
[Configuration of Digital Camera 100]
Hereinafter, a digital camera according to a first embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram for explaining a configuration of a digital camera according to the first embodiment of the present invention.

  In the digital camera 100 of the present embodiment, subject light enters from the left side of FIG. 1, a lens 10 (zoom lens or focus lens), a shutter 11, a diaphragm 12 (also referred to as an iris) for adjusting the amount of subject light, and After passing through the optical filter 13, an image is formed on the image sensor 14. The lens 10, the shutter 11, and the diaphragm 12 are controlled by a lens driving unit 15, a shutter driving unit 16, and a diaphragm driving unit 17, respectively.

  The lens driving unit 15 drives a zoom motor (not shown) according to an instruction transmitted from the CPU 18, thereby moving the zoom lens forward and backward along the optical axis to position the zoom lens at a predetermined position, thereby changing the focal length. In addition, the lens driving unit 15 drives a focus motor (not shown) according to an instruction transmitted from the CPU 18, thereby moving the focus lens forward and backward along the optical axis to position the focus lens at a predetermined position and changing the focal length. .

  The shutter drive unit 16 controls the shutter 11 by driving a shutter motor (not shown) in accordance with an instruction transmitted from the CPU 18. The aperture driving unit 17 drives an iris motor (not shown) according to an instruction transmitted from the CPU 18 to change the aperture amount of the aperture continuously or stepwise to change the aperture value.

  The optical filter 13 is, for example, a type of filter that controls the amount of transmitted light by controlling electrophoresis of nanoparticles. This type of optical filter is described in, for example, Japanese Patent Application Laid-Open No. 2006-72009.

  The imaging element 14 includes a photoelectric conversion element, and is configured by a color CCD having a predetermined color filter array (for example, a honeycomb array), for example. Light that has entered the light receiving surface of the image sensor 14 via the lens 10 or the like is converted into a signal charge in an amount corresponding to the amount of incident light by a photoelectric conversion element such as a photodiode arranged in a lattice pattern on the light receiving surface. The signal charges accumulated in the photoelectric conversion elements are read out in accordance with a timing signal applied from an image sensor driving unit 19 such as a timing generator (TG) and sequentially output from the image sensor 14 as a voltage signal (image signal). .

  The analog signal processing unit 20 includes a CDS circuit and an analog amplifier. The CDS circuit performs correlated double sampling processing on the CCD output signal based on the pulse applied from the image sensor driving unit 19, and the analog amplifier is output from the CDS circuit with a gain set according to the imaging sensitivity applied from the CPU 18. Amplify the image signal.

  The A / D converter 21 converts the analog image signal output from the analog signal processing unit 20 into a digital image signal (R, G, B RAW data). The converted digital image signal is stored in a storage means (not shown) such as a RAM.

  The digital signal processing unit 22 embeds an image signal stored in a storage unit such as a RAM and generates a YC signal composed of a luminance signal and a color difference signal. For example, the digital signal processing unit 22 performs offset processing for aligning the black levels of the R, G, and B signals, gain correction of the R, G, and B signals using the WB gains Rg, Gg, and Bg corresponding to the R, G, and B signals, WB correction by this gain correction, R, G, B signal tone conversion (gamma correction), and R, G, B signal spatial displacements associated with the color filter array of a single CCD are interpolated to obtain R, G, Processing for converting B signal into simultaneous expression, processing for converting R, G, B signals into luminance signal Y and color difference signals Cr, Cb, processing for emphasizing the contour portion (a portion where the luminance change is large), color difference The signal Cr, Cb is subjected to necessary matrix conversion to perform processing for realizing good color reproducibility.

  The display unit 23 is an image display device having a display surface such as a liquid crystal display. When a so-called through image is displayed on the display unit 23, images are continuously captured by the image sensor 14, and a plurality of frames obtained are obtained. The YC signal is generated by continuously processing the image signal of the image. The generated YC signal is added to the D / A converter 24, converted into a signal format for display, and output to the display 23. As a result, a through image is displayed on the display unit 23.

  When an image is recorded, the image is picked up in accordance with a shooting command from the operation unit 34, and the obtained image signal is processed to generate a YC signal. The generated YC signal is converted into predetermined compressed image data (for example, JPEG) by the recording / playback signal processing unit 25 and the like, and then stored in the external memory 26.

  The AE calculator 27 calculates a physical quantity necessary for AE control from the input image signal. For example, as a physical quantity necessary for AE control, one screen is divided into a plurality of areas (for example, 8 × 8), and an integrated value of R, G, and B image signals is calculated for each divided area. Then, the AE calculation unit 27 detects the brightness of the subject based on the calculated integrated value, the aperture value at the time of image signal acquisition, and the shirt speed, and obtains the EV value, and the program line based on the obtained EV value. Set the exposure based on the surroundings. The shutter drive unit 16 and the aperture drive unit 17 drive the shutter motor (not shown) and the iris motor (not shown) to control the shutter 11 and the aperture 12 so that the set exposure can be obtained.

  The AF calculation unit 28 calculates a physical quantity necessary for AF control from the input image signal. For example, the AF calculation unit 28 calculates an AF evaluation value indicating the sharpness of the image from the input image signal. The lens driving unit 15 drives a focus motor (not shown) so that the calculated AF evaluation value becomes a maximum, and moves the focus lens back and forth to a predetermined position.

  The face image extraction unit 29 performs face image extraction processing for extracting an image (face image) including a face from a captured image (for example, a digital image signal stored in a storage unit such as a RAM). For example, color data for face recognition is stored in advance, the color data of the entire area of the photographed image is compared with the color data for face recognition, and the matching portion is recognized as a person's face. A region including the face is cut out (see Japanese Patent Laid-Open No. 10-334213). In addition, various face image extraction processes can be applied. If a plurality of faces are imprinted on the captured image, a plurality of face images are extracted.

  The feature point extraction unit 30 performs a feature point extraction process for extracting a feature point of a face included in the face image from the face image extracted by the face image extraction unit 29. For example, the degree of coincidence between the face image and the data in the shape database unit is calculated, and feature points of the image are extracted based on the degree of coincidence and the shape data (see Japanese Patent Laid-Open No. 05-197793). In addition, various feature point extraction processes such as those described in Japanese Patent Application Laid-Open No. 08-077334 can be applied.

  The feature point change detection unit 31 performs a feature point change detection process for detecting a change over time of the facial feature points extracted by the feature point extraction unit 30.

  The person determination unit 32 determines whether the face included in the face image extracted by the face image extraction unit 29 is a person based on the temporal change in the feature points of the face detected by the feature point change detection unit 31. That is, a process of determining whether the person is actually in the place or a poster (or a picture or a figure) is performed.

  The person extraction unit 33 performs processing for associating the facial feature points extracted by the feature point extraction unit 30 with the results determined by the person determination unit 32 and storing them in a storage unit such as a RAM.

[Operation of Digital Camera 100]
Next, the operation (part 1) of the digital camera 100 having the above-described configuration will be described with reference to the drawings.

  FIG. 2 is a flowchart for explaining the operation (part 1) of the digital camera of the first embodiment.

  The following processing is processing for determining whether or not the face included in the face image is a person based on the change over time of the facial feature points extracted from the face image. The CPU 18 is mainly read into the RAM or the like. This is realized by executing the predetermined program.

  When the power of the digital camera 100 is turned on, continuous shooting is performed by the image sensor 14, and image data is generated from the output signal of the image sensor 14 and stored in storage means such as a RAM (step S10).

  The face image extraction unit 29 performs a face image extraction process for extracting a face image including a face from image data stored in a storage unit such as a RAM (step S11). Next, as shown in FIGS. 3, 4, and 5, the feature point extraction unit 30 performs facial feature points (for example, eyebrows, etc.) included in the face image from the face image extracted by the face image extraction unit 29. A process of extracting eyes, lips, or teeth) is performed (step S12). The processing from step S10 to step S12 is repeated within a predetermined period or for a predetermined number of frames.

  Next, the feature point change detection unit 31 performs a feature point change detection process for detecting a change over time of the facial feature points extracted by the feature point extraction unit 30 (step S13). For example, the degree of coincidence of feature points between the same face images between the frames is calculated, and when the degree of coincidence exceeds a predetermined threshold, it is detected that the feature points have changed over time.

  When the feature point change detection unit 31 detects that the feature point has changed over time (the left subject in FIGS. 4 and 5) (step S14: Yes), the person determination unit 32 (step S14: Yes). It is determined that the face included in the face image extracted by (1) is a person, that is, a person who is actually in that place (step S15). On the other hand, when the feature point change detection unit 31 does not detect the change of the feature point over time (the subject on the right side of FIGS. 4 and 5) (step S14: No), the face image extracted by the face image extraction unit 29 is added. It is determined that the included face is not a person, that is, a poster (or painting or figure) (step S16).

  The person extraction unit 33 includes a facial feature point (or face image) extracted by the feature point extraction unit 30 and a result determined by the person determination unit 32 (an identifier for identifying that the face image is a person, Alternatively, an identifier or the like for identifying that the face image is not a person is associated and stored in a storage means such as a RAM.

  As described above, in this operation, the facial feature points included in the face image are extracted from the image data (steps S11 and S12), and the temporal change of the facial feature points is detected (step S13). If a temporal change in the facial feature point is detected (step S14: Yes), it is determined that the face included in the face image is a person, that is, a person who is actually in that place ( Step S15). On the other hand, when a change with time of the facial feature point is not detected (step S14: No), it is determined that the face included in the face image is not a person, that is, a poster (or a picture or a figure) (step S16). ).

Therefore, when a subject including a poster other than a person is photographed by the digital camera 100, it is possible to prevent the poster or the like from being recognized as a person.
[Operation of Digital Camera 100 Part 2]
Next, another operation (part 2) of the digital camera 100 configured as described above will be described.

  The following processing, for example, corrects the amount of camera shake even if the facial feature point extracted due to camera shake or the like changes (for example, changes from frame A shown in FIG. 6 to frame B shown in FIG. 7). This is a process for determining whether or not the face included in the face image is a person based on the temporal change of the facial feature points, and is executed, for example, between steps S12 and S13 in FIG.

  The feature point extraction unit 30 extracts feature points of the contour of the subject for each of a predetermined number of frames (for example, the frame A shown in FIG. 6 and the frame B shown in FIG. 7). For example, A1, B1... In FIG. 8A, A2, B2... In FIG. 8B, A′1, B′1. Then, A′2, B′2,... Are extracted as feature points of the contour of the subject. Then, each frame is rotated by a predetermined amount or shifted up and down and left and right so that the same feature points between the frames match.

  Thereafter, by performing the processing from step S13 onward, when a subject including a poster other than a person is photographed by the digital camera 100, it is possible to prevent the poster or the like from being recognized as a person. Since it is possible to remove changes in the extracted facial feature points due to camera shake or the like, it is possible to reduce erroneous determination due to camera shake or the like.

[Operation of Digital Camera 100]
Next, another operation (part 3) of the digital camera 100 will be described with reference to the drawings.

  FIG. 10 is a flowchart for explaining another operation (No. 3) of the digital camera 100.

  The following processing determines whether or not the face included in the face image is a person based on the temporal change of the facial feature points extracted from the face image only when the area of the face image is a certain value or more. . This process is realized mainly by the CPU 18 executing a predetermined program read into the RAM or the like.

  When the power of the digital camera 100 is turned on, continuous imaging is performed by the image sensor 14, and image data is generated from the output signal of the image sensor 14 and stored in storage means such as a RAM (step S20).

  The face image extraction unit 29 performs face image extraction processing for extracting a face image including a face from image data stored in a storage unit such as a RAM (step S21). Next, as shown in FIGS. 3, 4, and 5, the feature point extraction unit 30 performs facial feature points (for example, eyebrows, etc.) included in the face image from the face image extracted by the face image extraction unit 29. A process of extracting eyes, lips, or teeth is performed (step S22). The processing from step S20 to step S22 is repeated within a predetermined period or for a predetermined number of frames.

  Next, the area of the face image extracted by the face image extraction unit 29 is obtained, and it is determined whether or not the obtained area is a certain value or more (step S23). For example, as shown in FIG. 11, the area of the face image corresponds to the face from the method of obtaining the height and width from the feature points of the contour and multiplying the height and width, and the luminance value of each image of the face image. For example, a method of identifying a pixel to be determined and obtaining the pixel based on the pixel can be considered. Alternatively, it may be obtained by multiplying the height and width of the face image itself.

  As a result, if the area is greater than or equal to a certain value (step S23: greater than or equal to a certain value), the feature point change detecting unit 31 detects a change over time of the facial feature points extracted by the feature point extracting unit 30. Point change detection processing is performed (step S24). For example, the degree of coincidence of feature points between the same face images between the frames is calculated, and when the degree of coincidence exceeds a predetermined threshold, it is detected that the feature points have changed over time.

  When the feature point change detection unit 31 detects that the feature point has changed over time (the left subject in FIGS. 4 and 5) (step S25: Yes), the person determination unit 32 (step S25: Yes). It is determined that the face included in the face image extracted by (1) is a person, that is, a person who is actually in the place (step S26). On the other hand, when the feature point change detection unit 31 detects that the feature point has not changed with time (the right subject in FIGS. 4 and 5) (step S25: No), the face extracted by the face image extraction unit 29 It is determined that the face included in the image is not a person, that is, a poster (or a picture or a figure) (step S27).

  On the other hand, if the area is less than a certain value (step S23: less than a certain value), it is determined that the face included in the face image extracted by the face image extraction unit 29 is not a person, that is, a poster (or a picture or a figure). (Step S27).

  The person extraction unit 33 includes a facial feature point (or face image) extracted by the feature point extraction unit 30 and a result determined by the person determination unit 32 (an identifier for identifying that the face image is a person, Alternatively, an identifier or the like for identifying that the face image is not a person is associated and stored in a storage means such as a RAM.

  As described above, in this operation, only when the area of the face image is greater than or equal to a certain value (step S23: greater than or equal to a certain value), the temporal feature point extracted by the feature point extraction unit 30 is changed over time. A feature point change detection process to be detected is performed (step S24), and it is determined whether or not the face included in the face image is a person based on the temporal change of the facial feature points extracted from the face image (step S24). S25).

  Therefore, it is possible to reduce misjudgment or useless processing due to the area of the face image not exceeding a certain value.

[Operation of Digital Camera 100 Part 4]
Next, another operation (part 4) of the digital camera 100 will be described with reference to the drawings.

  FIG. 12 is a flowchart for explaining another operation (No. 4) of the digital camera 100.

  The following processing is performed only when the focus position is at a predetermined position on the wide-angle side, based on the temporal change of the facial feature points extracted from the face image, whether or not the face included in the face image is a person. Determine whether. This process is realized mainly by the CPU 18 executing a predetermined program read into the RAM or the like.

  When the power of the digital camera 100 is turned on, continuous shooting is performed by the image sensor 14, image data is generated from the output signal of the image sensor 14, and stored in storage means such as a RAM (step S30).

  The face image extraction unit 29 performs a face image extraction process for extracting a face image including a face from image data stored in a storage unit such as a RAM (step S31). Next, as shown in FIGS. 3, 4, and 5, the feature point extraction unit 30 performs facial feature points (for example, eyebrows, etc.) included in the face image from the face image extracted by the face image extraction unit 29. A process of extracting eyes, lips, or teeth) is performed (step S32). The processing from step S30 to step S32 is repeated within a predetermined period or for a predetermined number of frames.

  Next, the position of the zoom lens is obtained, and it is determined whether or not the obtained position is located on the telephoto side from a certain value (step S33).

  As a result, if the obtained position is located on the telephoto side from the constant value (step S33: telephoto from the constant value), the feature point change detecting unit 31 extracts the facial features extracted by the feature point extracting unit 30. A feature point change detection process for detecting a change of the point with time is performed (step S34). For example, the degree of coincidence of feature points between the same face images between the frames is calculated, and when the degree of coincidence exceeds a predetermined threshold, it is detected that the feature points have changed over time.

  When the feature point change detection unit 31 detects that the feature point has changed with time (left subject in FIGS. 4 and 5) (step S35: Yes), the person determination unit 32 (step S35: Yes). It is determined that the face included in the face image extracted by (1) is a person, that is, a person who is actually in that place (step S36). On the other hand, when the feature point change detection unit 31 detects that the feature point has not changed with time (the right subject in FIGS. 4 and 5) (step S35: No), the face extracted by the face image extraction unit 29 It is determined that the face included in the image is not a person, that is, a poster (or painting or figure) (step S37).

  On the other hand, if the obtained position is located on the wide angle side from the certain value (step S33: wide angle from the certain value), the face included in the face image extracted by the face image extracting unit 29 is not a person, that is, Then, it is determined as a poster (or a picture or a figure) (step S37).

  The person extraction unit 33 includes a facial feature point (or face image) extracted by the feature point extraction unit 30 and a result determined by the person determination unit 32 (an identifier for identifying that the face image is a person, Alternatively, an identifier or the like for identifying that the face image is not a person is associated and stored in a storage means such as a RAM.

  As described above, in this operation, the facial features extracted by the feature point extraction unit 30 only when the position of the zoom lens is located on the telephoto side from a certain value (step S33: telephoto from a certain value). Whether or not the face included in the face image is a person based on the change over time of the feature points of the face extracted from the face image is performed (step S24). It is determined whether or not (step S35).

  As a result, an image with less focus shift can be processed, and erroneous determination can be reduced. Further, the processing can be reduced in weight.

  The zoom position determination step S33 may be provided at the position shown in FIG.

[Operation of digital camera 100]
Next, another operation (part 5) of the digital camera 100 will be described with reference to the drawings.

  FIG. 14 is a flowchart for explaining another operation (No. 5) of the digital camera 100.

  The following processing includes the face feature point (or face image) extracted by the feature point extraction unit 30 and the result determined by the person determination unit 32 (identifier for identifying that the face image is a person, or , An identifier for identifying that the face image is not a person) and the like are stored in storage means such as a RAM, and the newly extracted facial feature points are stored in the storage means. When the determination is made, the determination result stored in association with the facial feature point is read from the storage means. This process is realized mainly by the CPU 18 executing a predetermined program read into the RAM or the like.

  When the power of the digital camera 100 is turned on, continuous shooting is performed by the image sensor 14, image data is generated from the output signal of the image sensor 14, and stored in storage means such as a RAM (step S40).

  The face image extraction unit 29 performs a face image extraction process for extracting a face image including a face from image data stored in a storage unit such as a RAM (step S41). Next, as shown in FIGS. 3, 4, and 5, the feature point extraction unit 30 performs facial feature points (for example, eyebrows, etc.) included in the face image from the face image extracted by the face image extraction unit 29. A process of extracting eyes, lips, or teeth is performed (step S42). The processing from step S40 to step S42 is repeated within a predetermined period or for a predetermined number of frames.

  Next, it collates with the face image in the storage means (step S43). That is, it is determined whether or not the facial feature points (newly extracted facial feature points) extracted by the feature point extraction unit 30 are stored in a storage unit such as a RAM.

  As a result, if the newly extracted facial feature point is not stored in the storage means (step S43: does not match the image in the memory), the feature point change detection unit 31 uses the feature point extraction unit 30. A feature point change detection process is performed to detect a change with time of the extracted facial feature points (step S44). For example, the degree of coincidence of feature points between the same face images between the frames is calculated, and when the degree of coincidence exceeds a predetermined threshold, it is detected that the feature points have changed over time.

  When the feature point change detection unit 31 detects that the feature point has changed with time (left subject in FIGS. 4 and 5) (step S45: Yes), the person determination unit 32 (step S45: Yes). It is determined that the face included in the face image extracted by (1) is a person, that is, a person who is actually in that place (step S46). On the other hand, when the feature point change detection unit 31 detects that the feature point has not changed over time (the right subject in FIGS. 4 and 5) (step S45: No), the face extracted by the face image extraction unit 29 It is determined that the face included in the image is not a person, that is, a poster (or a picture or a figure) (step S48).

  Then, the person extraction unit 33 identifies the facial feature points (or face images) extracted by the feature point extraction unit 30 and the results determined by the person determination unit 32 (for identifying that the face image is a person). An identifier or an identifier for identifying that the face image is not a person) is stored in a storage means such as a RAM.

  On the other hand, if the newly extracted facial feature point is stored in the storage means (step S43: coincides with the image in the memory), it is stored in correspondence with the facial feature point from the storage means. Read the judgment result.

  If the read determination result indicates that the person is not a person (step S47: poster), the face included in the face image extracted by the face image extraction unit 29 is not a person, that is, a poster (or It is determined that it is a painting or a figure (step S48). On the other hand, if the determination result indicates that the person is a person (step S47: person), the face included in the face image extracted by the face image extraction unit 29 is a person, that is, the actual place (Step S46).

  As described above, in this operation, when it is determined that the newly extracted facial feature points are stored in the storage unit (step S43: match the image in the memory), the extracted face Since it is not necessary to detect the change of the feature point over time (the process of steps S44 and S45 is not necessary), it is possible to determine whether or not the person is a person very quickly.

[Second Embodiment]
[Configuration of Digital Camera 200]
Next, a digital camera according to a second embodiment of the present invention will be described with reference to the drawings. FIG. 17 is a block diagram for explaining a configuration of a digital camera according to the second embodiment of the present invention.

  In addition to the configuration of the digital camera 100 of the first embodiment, the digital camera 200 of the present embodiment further includes a configuration 40 for obtaining a distance image (light emitting element driving unit 41, light emitting element 42, optical filter 43, image sensor driving). Unit 44, light receiving element 45, distance calculating unit 46, etc.) and image collating unit 47.

  The light emitting element driving unit 41 causes the light emitting element 42 to emit light at a predetermined timing in accordance with an instruction transmitted from the CPU 18 to irradiate the subject with light, and the reflected light from the subject is captured through the optical filter 43 and the like, and the imaging element driving unit 44. And a distance image having a pixel value as a distance value obtained by converting a phase difference of light from light emission to light reception into a distance to a subject by a distance calculation unit 46 or the like (see, for example, Japanese Patent Laid-Open No. 2006). -172210).

  The image collation unit 47 collates the face image extracted by the face image extraction unit 29 with the distance image, and calculates the distance to each of the facial feature points (for example, eyes and nose) extracted by the feature point extraction unit 30. To do.

  The person determination unit 32 determines whether the face included in the face image extracted by the face image extraction unit 29 is a person based on the distance to each feature point in the face image calculated by the image matching unit 47, that is, Then, a process of determining whether the person is actually in the place or a poster (or a picture or a figure) is performed.

  Since other configurations are the same as those in the first embodiment, the same reference numerals are given and description thereof is omitted.

  Next, the operation of the digital camera 200 having the above configuration will be described with reference to the drawings.

[Operation of Digital Camera 200]
FIG. 18 is a flowchart for explaining the operation of the digital camera of the second embodiment.

  The following processing is processing for determining whether or not the face included in the face image is a person based on the distance to each feature point in the face image. Mainly, the CPU 18 reads a predetermined program read into the RAM or the like. It is realized by executing.

  When the power of the digital camera 200 is turned on, continuous shooting is performed by the image sensor 14, and image data is generated from the output signal of the image sensor 14 and stored in storage means such as a RAM (step S50).

  The face image extraction unit 29 performs a face image extraction process for extracting an image including a face from image data stored in a storage unit such as a RAM (step S51).

  Next, the light emitting element 42 is caused to emit light at a predetermined timing to irradiate the subject with light (step S52), and the reflected light from the subject is received by the light receiving element 45 through the optical filter 43 or the like (step S53) to calculate the distance. A distance image having a pixel value as a distance value obtained by converting the phase difference of light from light emission to light reception into the distance to the subject by the unit 46 or the like is obtained (step S54).

  Next, as shown in FIGS. 3, 4, and 5, the feature point extraction unit 30 extracts facial feature points (right eye, left eye, and right eye) included in the face image from the face image extracted by the face image extraction unit 29. A process of extracting the nose is performed (step S55).

  Then, the image collating unit 47 collates the face image extracted by the face image extracting unit 29 with the distance image, and each of the right eye, the left eye, and the nose which are the facial feature points extracted by the feature point extracting unit 30 is obtained. Distances L1, L2, and L3 are obtained (step S56).

  The person determination unit 32 calculates whether or not these distances L1, L2, and L3 satisfy a person condition (step S57). For example, if the condition of the person (L1 + L2) / 2 = L3 is calculated and the condition of the person is satisfied (step S58: Yes), the face included in the face image extracted by the face image extraction unit 29 is a person. It is determined that there is a person who is actually at the place (step S59). On the other hand, if the condition of the person is not satisfied (step S58: No), it is determined that the face included in the face image extracted by the face image extraction unit 29 is not a person, that is, a poster (or a picture or a figure). (Step S60).

  The person extraction unit 33 includes a facial feature point (or face image) extracted by the feature point extraction unit 30 and a result determined by the person determination unit 32 (an identifier for identifying that the face image is a person, Alternatively, an identifier or the like for identifying that the face image is not a person is associated and stored in a storage means such as a RAM.

  As described above, in this operation, the distances L1, L2, and L3 to the right eye, left eye, and nose, which are the extracted facial feature points, are obtained (step S56), and these distances L1, L2, and L3 are obtained. It is calculated whether or not the condition of the person is satisfied (step S57). If the condition of the person is satisfied (step S58: Yes), it is determined that the face included in the face image is a person, that is, a person who is actually in the place (step S59). On the other hand, if the condition of the person is not satisfied (step S58: No), it is determined that the face included in the face image is not a person, that is, a poster (or a picture or a figure) (step S60).

  Therefore, when a subject including a poster other than a person is photographed by the digital camera 100, it is possible to prevent the poster or the like from being recognized as a person.

  The person condition is not limited to (L1 + L2) / 2 = L3. When it is assumed that the error ΔL is included in the distance measurement, the condition of the person may be (L1 + L2) / 2> L3 + ΔL. Furthermore, a condition other than a person (L1 + L2) / 2 <L3 + ΔL may be employed.

  Alternatively, L1 <L3 <L2 or L1> L3> L2 may be adopted as a condition other than the person, and a case where these conditions are not satisfied may be adopted as the condition of the person. In addition, various conditions can be adopted as the condition of the person.

[Third embodiment]
[Configuration of Digital Camera 300]
Next, a digital camera according to a third embodiment of the present invention will be described with reference to the drawings. FIG. 19 is a block diagram for explaining a configuration of a digital camera according to the third embodiment of the present invention.

  In addition to the configuration of the digital camera 200 of the second embodiment, the digital camera 300 of the present embodiment further includes a zoom magnification calculation unit 50 for removing the influence of zoom.

  Since other configurations are the same as those in the second embodiment, the same reference numerals are given and description thereof is omitted.

[Operation of Digital Camera 300 Part 1]
Next, an operation (part 1) of the digital camera 300 having the above-described configuration will be described with reference to the drawings.

  FIG. 20 is a flowchart for explaining the operation (part 1) of the digital camera of the third embodiment.

  The following processing is mainly realized by the CPU 18 executing a predetermined program read into the RAM or the like.

  When the power of the digital camera 200 is turned on, continuous imaging is performed by the image sensor 14, image data is generated from the output signal of the image sensor 14, and stored in a storage means such as a RAM (step S70).

  The face image extraction unit 29 performs face image extraction processing for extracting an image including a face from image data stored in storage means such as a RAM (step S71).

  Next, the light emitting element 42 is caused to emit light at a predetermined timing to irradiate the subject with light (step S72), and the reflected light from the subject is received by the light receiving element 45 through the optical filter 43 or the like (step S73) to calculate the distance. A distance image having a pixel value as a distance value obtained by converting the phase difference of light from light emission to light reception into the distance to the subject by the unit 46 or the like is obtained (step S74).

  Next, as shown in FIGS. 3, 4, and 5, the feature point extraction unit 30 extracts facial feature points (right eye, left eye, and right eye) included in the face image from the face image extracted by the face image extraction unit 29. A process of extracting the nose is performed (step S75).

  Then, the zoom magnification calculator 50 corrects the influence of the zoom magnification on the image area corresponding to the feature point of the face image (step S76). This is a process for removing the influence of zoom in order to determine the size of the feature point. It is determined from the size and distance of the feature point from which the influence of zoom is removed whether or not it is appropriate as a person.

  This determination method will be described with reference to FIGS.

  FIG. 21 shows a state where a person is located in front of the poster. FIG. 22 shows the positional relationship between the poster and person shown in FIG. 21 and the digital camera 300. FIG. 23 shows the relationship between “distance from the digital camera 300” and “distance between feature points on the image”.

  As shown in FIG. 21, the vertexes of the right eye, left eye, and nose are adopted as feature points of the face image. For example, each face is represented by T1, T2, T3, and T'1, T'2, T'3. . Further, the centroids of T1, T2, and T3 and the centroids of T′1, T′2, and T′3 are represented by T4 and T′4, respectively. Further, as shown in FIG. 22, the distances from the digital camera 300 to the centroids T4 and T′4 are represented by L4 and L′ 4, respectively. As shown in FIG. 21, for example, the distances between the right eye and the nose vertices are represented as L5 and L5 ′ as “distances between feature points on the image”.

  23, a coordinate system of “distance from the digital camera 300” and “distance between feature points on the image” shown in FIG. 23, and a person area and a person set on this coordinate system. A table or expression representing an area other than (portrait area in the figure) is stored.

  The zoom magnification calculation unit 50 collates the face image extracted by the face image extraction unit 29 with the distance image and obtains distances L4, L5, L′ 4, and L′ 5 (step S77).

  The person determination unit 32 determines whether or not these distances L4, L5, L′ 4, and L′ 5 satisfy the person condition (step S78). That is, it is determined whether or not the position of the coordinate system corresponding to the distances L4 and L5 (or the position of the coordinate system corresponding to L′ 4 and L′ 5) corresponds to the person area. As a result, if the position of the coordinate system corresponding to the distances L4 and L5 (or the position of the coordinate system corresponding to the distances L′ 4 and L′ 5) corresponds to the person area (the condition of the person is satisfied) (step S78: Yes), it is determined that the face included in the face image extracted by the face image extraction unit 29 is a person, that is, a person who is actually in that place (step S79). On the other hand, if the position of the coordinate system corresponding to the distances L4 and L5 (or the position of the coordinate system corresponding to the distances L′ 4 and L′ 5) does not correspond to the person area (does not satisfy the condition of the person) (step) S78: No), it is determined that the face included in the face image extracted by the face image extraction unit 29 is not a person, that is, a poster (or a picture or a figure) (step S80).

  The person extraction unit 33 includes a facial feature point (or face image) extracted by the feature point extraction unit 30 and a result determined by the person determination unit 32 (an identifier for identifying that the face image is a person, Alternatively, an identifier or the like for identifying that the face image is not a person is associated and stored in a storage means such as a RAM.

  As described above, in this operation, it is determined whether or not the distances L4, L5, L′ 4, and L′ 5 satisfy the condition of the person (step S78). If the condition of the person is satisfied (step S78: Yes), it is determined that the face included in the face image is a person, that is, a person who is actually in the place (step S79). On the other hand, if the condition of the person is not satisfied (step S78: No), it is determined that the face included in the face image is not a person, that is, a poster (or a picture or a figure) (step S80).

  Therefore, when a subject including a poster other than a person is photographed by the digital camera 100, it is possible to prevent the poster or the like from being recognized as a person.

  Further, it is possible to determine whether each face image is a person based on the face length and distance information from which the influence of zoom is removed.

  Note that the probability of erroneous determination can be reduced by performing the above processing on at least two sets of feature points.

[Operation of Digital Camera 300 Part 2]
Next, another operation (part 2) of the digital camera 300 configured as described above will be described with reference to the drawings.

  FIG. 24 is a flowchart for explaining another operation (No. 2) of the digital camera 300 according to the third embodiment.

  The following processing is mainly realized by the CPU 18 executing a predetermined program read into the RAM or the like.

  When the power of the digital camera 300 is turned on, continuous shooting is performed by the image sensor 14, image data is generated from the output signal of the image sensor 14, and stored in storage means such as a RAM (step S90).

  The face image extraction unit 29 performs face image extraction processing for extracting an image including a face from image data stored in storage means such as a RAM (step S91).

  Next, the light emitting element 42 is caused to emit light at a predetermined timing to irradiate the subject with light (step S92), and the reflected light from the subject is received by the light receiving element 45 through the optical filter 43 or the like (step S93) to calculate the distance. A distance image having a pixel value as a distance value obtained by converting the phase difference of light from light emission to light reception into the distance to the subject by the unit 46 or the like is obtained (step S94).

  Next, as shown in FIGS. 3, 4, and 5, the feature point extraction unit 30 extracts facial feature points (right eye, left eye, and right eye) included in the face image from the face image extracted by the face image extraction unit 29. A process of extracting the nose is performed (step S95).

  Then, the image collating unit 47 collates the face image extracted by the face image extracting unit 29 with the distance image, and each of the right eye, the left eye, and the nose which are the facial feature points extracted by the feature point extracting unit 30 is obtained. Distances L1, L2, and L3 are obtained (step S96).

  The person determination unit 32 calculates whether or not these distances L1, L2, and L3 satisfy the condition of the person (step S97). For example, if the condition of the person (L1 + L2) / 2 = L3 is calculated and the condition of the person is satisfied (step S98: Yes), the face included in the face image extracted by the face image extraction unit 29 is a person. It is determined that there is a person who is actually at the place (step S103). On the other hand, if the condition of the person is not satisfied (step S98: No), the following process is further performed.

  In other words, the zoom magnification calculator 50 corrects the influence of the zoom magnification on the image area corresponding to the feature point of the face image (step S99). This is a process for removing the influence of zoom in order to determine the size of the feature point. It is determined from the size and distance of the feature point from which the influence of zoom is removed whether or not it is appropriate as a person. This determination method has already been described (see step S76).

  The zoom magnification calculator 50 collates the face image extracted by the face image extractor 29 with the distance image, and obtains distances L4, L5, L′ 4, and L′ 5 (step S100).

  The person determination unit 32 determines whether or not these distances L4, L5, L′ 4, and L′ 5 satisfy a person condition (step S101). That is, it is determined whether or not the position of the coordinate system corresponding to the distances L4 and L5 (or the position of the coordinate system corresponding to the distances L′ 4 and L′ 5) corresponds to the person area. As a result, if the position of the coordinate system corresponding to the distances L4 and L5 (or the position of the coordinate system corresponding to the distances L′ 4 and L′ 5) corresponds to the person area (the condition of the person is satisfied) (step S101: Yes), it is determined that the face included in the face image extracted by the face image extraction unit 29 is a person, that is, a person who is actually in that place (step S103). On the other hand, if the position of the coordinate system corresponding to the distances L4 and L5 (or the position of the coordinate system corresponding to the distances L′ 4 and L′ 5) does not correspond to the person area (does not satisfy the condition of the person) (step) S101: No), it is determined that the face included in the face image extracted by the face image extraction unit 29 is not a person, that is, a poster (or a picture or a figure) (step S102).

  The person extraction unit 33 includes a facial feature point (or face image) extracted by the feature point extraction unit 30 and a result determined by the person determination unit 32 (an identifier for identifying that the face image is a person, Alternatively, an identifier or the like for identifying that the face image is not a person is associated and stored in a storage means such as a RAM.

  Note that the processing of steps S96 to S98 may be performed after step S101.

[Operation of Digital Camera 300 Part 3]
Next, another operation (part 3) of the digital camera 300 configured as described above will be described with reference to the drawings.

  FIG. 25 is a flowchart for explaining another operation (part 3) of the digital camera according to the third embodiment.

  The following processing is mainly realized by the CPU 18 executing a predetermined program read into the RAM or the like.

  When the power of the digital camera 300 is turned on, continuous shooting is performed by the image sensor 14, and image data is generated from the output signal of the image sensor 14 and stored in storage means such as a RAM (step S110).

  The face image extraction unit 29 performs face image extraction processing for extracting an image including a face from image data stored in storage means such as a RAM (step S111).

  Next, the light emitting element 42 is caused to emit light at a predetermined timing to irradiate the subject with light (step S112), and the reflected light from the subject is received by the light receiving element 45 through the optical filter 43 or the like (step S113) to calculate the distance. A distance image having a pixel value as a distance value obtained by converting the phase difference of light from light emission to light reception into the distance to the subject is obtained by the unit 46 or the like (step S114).

  Next, as shown in FIGS. 3, 4, and 5, the feature point extraction unit 30 extracts facial feature points (right eye, left eye, and right eye) included in the face image from the face image extracted by the face image extraction unit 29. A process of extracting the nose is performed (step S115).

  Then, the zoom magnification calculator 50 corrects the influence of the zoom magnification on the image area corresponding to the feature point of the face image (step S116). This is a process for removing the influence of zoom in order to determine the size of the feature point. It is determined from the size and distance of the feature point from which the influence of zoom is removed whether or not it is appropriate as a person. This determination method has already been described (see step S76).

  The zoom magnification calculator 50 collates the face image extracted by the face image extractor 29 with the distance image, and obtains distances L4, L5, L′ 4, and L′ 5 (step S117).

  The person determination unit 32 determines whether or not these distances L4, L5, L′ 4, and L′ 5 satisfy the condition of the person (step S118). That is, it is determined whether or not the position of the coordinate system corresponding to the distances L4 and L5 (or the position of the coordinate system corresponding to the distances L′ 4 and L′ 5) corresponds to the person area. As a result, if the position of the coordinate system corresponding to the distances L4 and L5 (or the position of the coordinate system corresponding to the distances L′ 4 and L′ 5) does not correspond to the person area (does not satisfy the person condition) ( Step S118: No), it is determined that the face included in the face image extracted by the face image extraction unit 29 is not a person, that is, a poster (or a picture or a figure) (step S123).

  On the other hand, if the position of the coordinate system corresponding to the distances L4 and L5 (or the position of the coordinate system corresponding to the distances L′ 4 and L′ 5) corresponds to the person area (the condition of the person is satisfied) (step S118). : Yes), the following processing is further performed.

  The image collation unit 47 collates the face image extracted by the face image extraction unit 29 with the distance image, and distances L1 to the right eye, the left eye, and the nose that are the facial feature points extracted by the feature point extraction unit 30. , L2 and L3 are obtained (step S119).

  The person determination unit 32 calculates whether or not these distances L1, L2, and L3 satisfy the condition of the person. For example, the condition (L1 + L2) / 2 = L3 of the person is calculated (step S120), and if the condition of the person is satisfied (step S121: Yes), it is included in the face image extracted by the face image extraction unit 29. It is determined that the face is a person, that is, a person who is actually at the place (step S122). On the other hand, if the condition of the person is not satisfied (for example, the right subject in FIGS. 4 and 5) (step S121: No), the face included in the face image extracted by the face image extraction unit 29 is not a person. That is, it is determined as a poster (or painting or figure) (step S123).

  The person extraction unit 33 includes a facial feature point (or face image) extracted by the feature point extraction unit 30 and a result determined by the person determination unit 32 (an identifier for identifying that the face image is a person, Alternatively, an identifier or the like for identifying that the face image is not a person is associated and stored in a storage means such as a RAM.

  Note that the processing of steps S116 to S118 may be performed after step S121. Further, as shown in FIG. 26, the process of step S14 described in FIG. 2 may be performed instead of steps S119 to S121.

  As described above, when a subject including a poster other than a person is photographed by the digital cameras 100, 200, and 300 by the operations of the above-described embodiments, the poster is prevented from being recognized as a person. It becomes possible.

  When it is determined that the face included in the face image is a person, more appropriate shooting can be performed by weighting the AE evaluation or preferentially focusing on the person. It becomes possible.

  Or the display showing the determination result of whether it is a person can also be displayed on a display surface. For example, as shown in FIG. 16, a subject determined to be a person is displayed in a so-called through image surrounded by a dotted rectangular frame (for example, blue), and a subject determined not to be a person is displayed as a dotted rectangular frame (for example, red). ). Alternatively, only a subject determined to be a person may be surrounded and displayed by a rectangular frame.

  The digital camera 100 is provided with means for detecting the displacement of the camera itself, such as a gyroscope, an acceleration sensor, a GPS, and an electronic compass, thereby correcting the amount of movement of the digital camera 100 and detecting the displacement of the feature point of the subject. Can also be considered.

  The above embodiment is merely an example in all respects. The present invention is not construed as being limited to these descriptions. The present invention can be implemented in various other forms without departing from the spirit or main features thereof.

It is a block diagram for demonstrating the structure of the digital camera which is 1st embodiment of this invention. It is a flowchart for demonstrating the operation | movement (the 1) of the digital camera of 1st embodiment. It is a figure for demonstrating the feature point extracted from a face image. It is a figure for demonstrating the change of a feature point. It is a figure for demonstrating the change of a feature point. It is a figure for demonstrating the state to which the to-be-photographed object moved relatively with the flame | frame of this time, and the flame | frame immediately before because the position of a digital camera shifted | deviated. It is a figure for demonstrating the state to which the to-be-photographed object moved relatively with the flame | frame of this time, and the flame | frame immediately before because the position of a digital camera shifted | deviated. It is a figure for demonstrating the feature point extracted from a face image. It is a figure for demonstrating the feature point extracted from a face image. 12 is a flowchart for explaining another operation (part 3) of the digital camera 100; It is a figure for demonstrating how to obtain | require the area of a face image. 12 is a flowchart for explaining another operation (No. 4) of the digital camera. 12 is a flowchart for explaining a modification of another operation (part 4) of the digital camera. 12 is a flowchart for explaining another operation (No. 5) of the digital camera. A state in which the facial feature points extracted by the feature point extraction unit 30 and the results determined by the person determination unit 32 are stored in the internal memory (a storage unit such as a RAM) of the digital camera 100 is schematically shown. It is a figure for demonstrating. It is an example of the display showing the determination result of whether it is a person. It is a block diagram for demonstrating the structure of the digital camera which is 2nd embodiment of this invention. It is a flowchart for demonstrating operation | movement of the digital camera of 1st embodiment. It is a block diagram for demonstrating the structure of the digital camera which is 3rd embodiment of this invention. 6 is a flowchart for explaining another operation (part 1) of the digital camera 300; It is a figure for demonstrating the method to correct | amend the influence of zoom magnification to the image area corresponding to the feature point of a face image. It is a figure for demonstrating the method to correct | amend the influence of zoom magnification to the image area corresponding to the feature point of a face image. It is a figure for demonstrating the method to correct | amend the influence of zoom magnification to the image area corresponding to the feature point of a face image. It is a flowchart for demonstrating the operation | movement (the 2) of the digital camera of 3rd embodiment. It is a flowchart for demonstrating the operation | movement (the 3) of the digital camera of 3rd embodiment. It is a flowchart for demonstrating the modification of operation | movement (the 3) of the digital camera of 3rd embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Lens, 11 ... Shutter, 13 ... Optical filter, 14 ... Image sensor, 15 ... Lens drive part, 16 ... Shutter drive part, 17 ... Diaphragm drive part, 19 ... Image sensor drive part, 20 ... Analog signal processing part, DESCRIPTION OF SYMBOLS 21 ... A / D converter, 22 ... Digital signal processing part, 23 ... Display part, 24 ... D / A conversion part, 25 ... Recording / reproduction signal processing part, 26 ... External memory, 27 ... AE calculating part, 28 ... AF Arithmetic unit, 29 ... face image extraction unit, 30 ... feature point extraction unit, 31 ... feature point change detection unit, 32 ... person discrimination unit, 33 ... person extraction unit, 34 ... operation unit, 41 ... light emitting element drive unit, 42 DESCRIPTION OF SYMBOLS ... Light emitting element, 43 ... Optical filter, 44 ... Image pick-up element drive part, 45 ... Light receiving element, 46 ... Distance calculating part, 47 ... Image collation part, 50 ... Zoom magnification calculating part, 100, 200, 300 ... Digital camera

Claims (6)

An image sensor including a photoelectric conversion element;
Image data generating means for generating image data from an output signal of the image sensor;
A face image extracting means for extracting a face image including a face from the image data generated by the image data generating means;
Feature point extracting means for extracting a feature point of a face included in the face image from the face image extracted by the face image extracting means;
A feature point change detecting means for detecting a change over time of the facial feature points extracted by the feature point extracting means;
A person determination unit for determining whether or not a face included in the face image extracted by the face image extraction unit is a person based on a temporal change in the feature point of the face detected by the feature point change detection unit; ,
A digital camera comprising: The feature point extracting unit extracts a feature point of a face outline included in the face image from the face image extracted by the face image extracting unit;
The feature point change detecting means is a camera shake amount of the facial feature point extracted by the feature point extracting means based on a change in the feature point with time of the contour of the face extracted by the feature point extracting means. To correct
The person determination means determines whether or not the face included in the face image extracted by the face image extraction means is a person based on the change over time of the facial feature points corrected by the feature point change detection means. The digital camera according to claim 1, wherein the determination is made. Further comprising area detection means for detecting the area of the face image extracted by the face image extraction means;
The digital camera according to claim 1, wherein the feature point change detection unit and the person determination unit execute the processes when the area detected by the area detection unit exceeds a certain value. A focus unit that preferentially focuses on a face included in the face image extracted by the face image extraction unit;
2. The digital according to claim 1, wherein the feature point change detection unit and the determination unit execute the processes when a focus position by the focus unit is located at a predetermined position on the wide-angle side. camera. Storage means for storing the facial feature points extracted by the feature point extraction means in association with the results determined by the person determination means;
In-memory determination means for determining whether or not the feature points of the face newly extracted by the feature point extraction means are stored in the storage means;
When it is determined that the feature point of the face newly extracted by the feature point extraction unit is stored in the storage unit by the determination unit in the memory, the feature point is associated with the face feature point from the storage unit The digital camera according to claim 1, wherein the determination result stored in the step is read out.   The digital camera according to claim 1, further comprising a display unit that displays a display representing a determination result by the person determination unit on a display surface.

Images