JP5408288B2 - Electronic camera - Google Patents

Description

  The present invention relates to an electronic camera.

  Conventionally, a technique has been disclosed in which feature points of a person's face are extracted from a photographed image of the person's face, and face recognition is performed based on the degree of coincidence between face recognition data related to the person to be recognized and the above feature points. (See Patent Document 1). Also disclosed is an image processing apparatus that corrects a face image by estimating race, age, and gender from a captured image (see Patent Document 2).

JP-A-6-259534 JP 2004-246456 A

  By the way, in recent years, multi-functionalization of electronic cameras has been progressing more and more, and accordingly, the operation of electronic cameras tends to become complicated. Therefore, there is a demand for an electronic camera that can easily use a function intended by a user when photographing a person.

  The present invention is to solve the above-mentioned problems of the prior art. An object of the present invention is to provide an electronic camera that can detect a face of a subject at the time of photographing a person and can easily perform a process requested by a user based on characteristics of the face of the subject.

Electronic camera according to an embodiment of the present invention includes an imaging device that photographs a subject image, a memory, a face detection unit, and sure識部, and speakers for outputting sound to the outside, and a control unit. Memory, the registered data indicating the feature quantity of the face, the correspondence between the characteristic data relating to only locations or language out is recorded, correspondence between the sound indicated cues of the characteristic data and the photographing is recorded. The face detection unit detects the face area of the subject image based on the output of the image sensor, and extracts the feature amount of the face of the subject image from the face area . Sure識部, based on the registration data recorded in the feature amount and the memory of the face of the object image extracted by the face detecting unit recognizes an object image. Control unit on the basis of the registered data of the object image recognized by certified識部to output a sound indicating the cue shooting recorded in the memory from the speaker before shooting.

  According to the present invention, it is possible to detect a face of a subject at the time of photographing a person and easily perform a process requested by a user based on a feature of the face of the subject.

The block diagram which shows the structure of the electronic camera of 1st Embodiment. A flow chart explaining an example of photography operation of an electronic camera in a 1st embodiment Flow chart for explaining an example of photographing operation of the electronic camera in the second embodiment

(Description of the first embodiment)
FIG. 1 is a block diagram showing the configuration of the electronic camera of the first embodiment.

  The electronic camera includes a photographing lens 11, a lens driving unit 12, a diaphragm 13, a diaphragm driving unit 14, an image sensor 15, an analog processing unit 16, an A / D conversion unit 17, and a timing generator (TG) 18. A first memory 19, an image processing unit 20, a recording I / F 21, a display I / F 22, a monitor 23, an operation unit 24, a communication unit 25, an external I / F 26, a microphone 27, and a speaker. 28, a flash light emitting unit 29, an external light AF unit 30, a CPU 31, a second memory 32, and a system bus 33. The first memory 19, the image processing unit 20, the recording I / F 21, the display I / F 22, the communication unit 25, the external I / F 26 and the CPU 31 are connected via a system bus 33.

  The photographing lens 11 is composed of a plurality of lens groups including a zoom lens and a focus lens for adjusting a focus position. Each lens constituting the photographing lens 11 is moved in the optical axis direction by the lens driving unit 12. The input / output of the lens driving unit 12 is connected to the CPU 31.

  The diaphragm 13 adjusts the amount of incident light from the photographic lens 11. The aperture amount of the diaphragm 13 is adjusted by the diaphragm drive unit 14. The input / output of the aperture driving unit 14 is connected to the CPU 31.

  The image sensor 15 photoelectrically converts the light beam that has passed through the photographing lens 11 to generate an analog image signal of the subject image. In the photographing mode for photographing a subject, the image sensor 15 captures a recording image for recording at the time of shooting, and outputs a through image by thinning out at predetermined intervals even during standby for shooting. The through image data is used for various arithmetic processes by the CPU 31 and display on the monitor 23.

  The analog processing unit 16 is an analog front end circuit that performs analog signal processing on the output of the image sensor 15. The analog processing unit 16 has a CDS circuit and a gain circuit inside. The CDS circuit of the analog processing unit 16 reduces the noise component of the output of the image sensor 15 by correlated double sampling. The gain circuit of the analog processing unit 16 amplifies the gain of the input signal based on an instruction from the CPU 31. In the gain circuit described above, the imaging sensitivity corresponding to the ISO sensitivity can be adjusted.

  The A / D conversion unit 17 performs A / D conversion on the analog image signal output from the analog processing unit 16. The output of the A / D converter 17 is connected to the first memory 19.

  The TG 18 supplies timing pulses to the image sensor 15, the analog processing unit 16, and the A / D conversion unit 17 based on instructions from the CPU 31. The drive timing of the image sensor 15, the analog processing unit 16, and the A / D conversion unit 17 is controlled by the timing pulse of the TG 18.

  The first memory 19 is a buffer memory for temporarily recording image data in the pre-process and post-process of image processing described later.

  The image processing unit 20 performs various kinds of image processing on the digital image signal at the time of shooting in the above shooting mode to generate a shot image. The image processing unit 20 compresses the captured image data in the JPEG format. Furthermore, the image processing unit 20 generates a display image (view image) from the image signal of the through image according to an instruction from the CPU 31 in the above-described shooting mode.

  On the other hand, in an edit mode in which a captured image is reproduced, the image processing unit 20 executes a process of decompressing and restoring the compressed captured image data. The image processing unit 20 can also perform image processing on captured image data in the editing mode. The specific contents of the image processing by the image processing unit 20 will be described later.

  A connector for connecting the recording medium 34 is formed in the recording I / F 21. The recording I / F 21 executes writing / reading of the captured image data with respect to the recording medium 34 connected to the connector. The recording medium 34 is composed of a hard disk, a memory card incorporating a semiconductor memory, or the like. In FIG. 1, a memory card is illustrated as an example of the recording medium 34.

  The display I / F 22 controls the display on the monitor 23 based on an instruction from the CPU 31. The display I / F 22 has a video terminal for outputting an image to the outside. Therefore, in the electronic camera of the present embodiment, it is possible to display a reproduced image on an external display device (such as a television) via the video terminal of the display I / F 22.

  The monitor 23 displays various images according to instructions from the CPU 31 and the display I / F 22. The monitor 23 in this embodiment is a liquid crystal monitor. The monitor 23 displays a moving image of the view image during shooting standby in the shooting mode. The monitor 23 displays various information necessary for shooting (for example, the number of frames that can be shot, the position of the focus detection area, the frame display of the face area detected by the face detection process described later, etc.) by the on-screen function. ) Can be superimposed on the view image.

  Further, the monitor 23 can also display a playback image based on the captured image data, a menu screen that can be input in a GUI (Graphical User Interface) format, and the like (the illustration of the view image and the menu screen on the monitor 23 is omitted). To do).

  The operation unit 24 has a release button, an operation button, and the like. The release button of the operation unit 24 receives an instruction input for starting an exposure operation from the user. The operation buttons of the operation unit 24 accept input from the user such as an input on the menu screen or the like, an input for switching the photographing mode of the electronic camera, and the like.

  The communication unit 25 controls communication with an external device in accordance with a specification defined in the wireless LAN standard IEEE 802.11. For example, when transmitting captured image data to an external device, the communication unit 25 converts the captured image data into data that can be wirelessly communicated, and transmits the converted data to the external device.

  The external I / F 26 has a connection terminal of a serial communication standard such as USB (Universal Serial Bus). The external I / F 26 controls data transmission / reception with a computer or the like connected via the connection terminal in accordance with the communication standard.

  The microphone 27 is provided for recording external sound of the electronic camera. The speaker 28 is provided to output sound to the outside of the electronic camera during shooting. The microphone 27 and the speaker 28 are connected to the CPU 31.

  The flash light emitting unit 29 includes, for example, a xenon arc tube, a main capacitor that stores light emission energy, a reflector or lens member for efficiently irradiating the subject with flash light, a light emission control circuit, and the like.

  The external light AF unit 30 is a known passive focus detection device that performs focus detection using an optical system independent of the photographing lens 11. The external light AF unit 30 includes a pair of distance measuring lenses 30a, mirrors 30b and 30c, and an image sensor 30d. The subject light that has passed through the pair of distance measuring lenses 30a is detected as two images on the image sensor 30d after being reflected by the mirrors 30b and 30c.

  The CPU 31 controls the operation of each part of the electronic camera according to a sequence program stored in a ROM (not shown), and executes various calculations necessary for photographing (the contents will be described later). Further, the CPU 31 of the present embodiment functions as an AF control unit 31a, a face detection unit 31b, a face recognition unit 31c, an inference unit 31d, and a timer 31e for acquiring the shooting date and time.

  The AF control unit 31a executes a known AF calculation using a contrast detection method based on the through image data acquired from the image sensor 15. The AF control unit 31a calculates the defocus amount based on the two-image interval detected from the image sensor of the external light AF unit 30, and performs AF based on the defocus amount. In the first embodiment, in the normal shooting mode, the AF control unit 31a performs AF with the defocus amount of the external light AF unit 30, and then further performs contrast detection AF to perform focusing. The accuracy is improved.

  The face detection unit 31b extracts feature points from the data of the through image or the photographed image, and detects the face area of the subject, the size of the face, and the like. For example, the face detection unit 31b extracts a face region by a feature point extraction process described in Japanese Patent Application Laid-Open No. 2001-16573. Examples of the feature points include eyebrow, eye, nose, and lip end points, face contour points, head vertices, and chin lower end points.

  The face recognition unit 31c generates face recognition data based on the feature points detected by the face detection unit 31b. For example, the face recognition unit 31c generates face recognition data of a registered person to be recognized from the position of the detected feature point of the face, the size of the face part obtained from the feature point, the relative distance of each feature point, and the like.

  In addition, the face recognition unit 31c performs face recognition processing for determining whether or not a person's face in the shooting screen is a person's face in the face recognition data. Specifically, the face recognition unit 31c first calculates the position of the feature point of the photographed person's face, the size of each face part, the relative distance of each feature point, and the like based on the detected feature points of the face. Next, the face recognition unit 31c compares the calculation result with the face recognition data, and obtains the similarity between the face of the person in the face recognition data and the face of the photographed person. Then, the face recognition unit 31c determines that the photographed person matches the person in the face recognition data when the similarity is above the threshold.

  The inference unit 31d discriminates the race, age, sex, and the like by arithmetic processing using feature points that constitute the facial organ. In the estimation of the race, for example, the technique of the following document 1 can be applied, but not limited to this, a method of detecting the luminance distribution in the face image can be used. Moreover, in order to estimate age and sex, the method disclosed by the following literature 2 can be used, for example.

<Reference 1>: Gregory Shakhnarovich, Paul A. Viola, Babak Mohaddam; “A Unified Learning Framework for Reality Detection for Real Time Face Detection and Classification ”; Proceedings of the Fifth IEEE International Conference on Automatic Face and Gesture Recognition (Proceedings of the Fifth International) (on the Institute of Electrical and Electronics Engineers, E2); Masato Kawade; “Gender and Dating System Using Gabor Wavelet Transform and Support Vector Machine”; Proceedings of the 8th Symposium on Image Sensing Symposium; Image Sensing Technology Study Group; July 2002 Features Used in the Inference Processing Is mainly derived from the detection area of the face image, but is not limited to this. Some of the features of the image may also be included. As the feature amount, it is possible to detect the average and variance of the color and brightness of the face image, the intensity distribution, the color difference and brightness difference with the surrounding image, and the like. Further, these feature quantities can be applied to a predetermined arithmetic expression to obtain secondary feature quantities necessary for inference.

  The second memory 32 is connected to the CPU 31. In the second memory 32, face recognition data generated by the face recognition unit 31c is recorded. In the second memory 32, a group folder can be generated for each registered person (registered person), and face recognition data can be grouped for each registered person by the group folder. For example, in the group folder, a plurality of face recognition data with different face orientations, shooting conditions, etc. can be recorded for the same registered person.

  Further, in each group folder of the second memory 32, characteristic data indicating characteristics of the registered person and settings at the time of shooting can be recorded.

  The above characteristic data includes (1) name of registered person, (2) date of birth of registered person, (3) gender of registered person, (4) race of registered person, and (5) birth place of registered person. / Language, (6) Presence / absence of wearing glasses, (7) Setting of incidental display on captured image, (8) Display of characters. Moreover, the registered person's telephone number, address, mail address, nickname, etc. may be recorded as the characteristic data.

  Here, the setting of the auxiliary display in (7) above indicates whether or not the auxiliary display is combined with the captured image. This supplementary display includes cartoon characters, stamps such as letters and pictures, templates such as frames that decorate the periphery of the photographed image, and character information such as news on the photographing date. In addition, the setting of the incidental display includes setting of a position where the incidental display is synthesized in the image. The accompanying display data is recorded in the second memory 32. Note that the CPU 31 may add or update the incidental display data in the second memory 32 by, for example, downloading the data from the external server via the communication unit 25.

  The character display in (8) is a setting item related to characters when characters are combined with the captured image (such as display of the shooting date and the character information of the accompanying display described above). For example, the setting includes the font type, character size, character color, and character position to be combined in the image.

  In addition, the second memory 32 stores voice data of a call used for a signal at the time of shooting. As the audio data, call data corresponding to a plurality of languages are recorded. In addition, a single language has a plurality of patterns of shout data in standard words and dialects.

  Next, an example of the photographing operation of the electronic camera in the first embodiment will be described with reference to the flowchart of FIG. Here, in the following description, a case will be described in which the reasoning function of a person's age and gender is set to ON and the face recognition function is set to OFF.

  Step 101: The CPU 31 drives the imaging device 15 to start acquiring a through image. The imaging element 15 acquires an image signal of a through image by thinning out reading at predetermined intervals. The image processing unit 20 sequentially generates view images based on through-image data. The view image is displayed as a moving image on the monitor 23 at the time of shooting standby. Therefore, the user can perform framing for determining the shooting composition based on the view image on the monitor 23.

  Step 102: The face detection unit 31b of the CPU 31 performs a face detection process on the through image data to detect a face area in the shooting screen.

  Step 103: The CPU 31 determines whether or not a face area has been detected in the face detection process of S102. If a face area is detected (YES side), the process proceeds to S104. On the other hand, when the face area is not detected (NO side), the process proceeds to S107.

  Step 104: The inference unit 31d of the CPU 31 executes an inference process for estimating the age and sex of the main subject for the face area in S102. As an example, the CPU 31 sets a plurality of classes for each category of the subject's sex and age, and obtains the likelihood for each class. Then, the CPU 31 estimates the age and sex of the main subject based on the class with the highest likelihood.

  Here, when a plurality of face areas are detected in S102, the CPU 31 infers as a main subject either the face area that is the largest face size and is considered to be the closest side or the face area that is located at the center most. Execute the process.

  Step 105: The CPU 31 determines whether or not the main subject is a child from the age estimated in the inference process of S104. Specifically, the CPU 31 determines that the main subject is a child when the estimated age of the main subject is equal to or less than a threshold (for example, 6 to 10 years old or less). If the main subject is a child (YES side), the process proceeds to S108. On the other hand, if the main subject is not a child (NO side), the process proceeds to S106.

  Step 106: The CPU 31 determines whether or not the main subject is a woman from the gender estimated in the inference process of S104. When the main subject is a woman (YES side), the process proceeds to S109. On the other hand, if the main subject is male (NO side), the process proceeds to S107.

  Step 107: In this case, the CPU 31 shoots the subject in the normal shooting mode. In the normal shooting mode, the CPU 31 performs AF in response to half-pressing of the release button, and executes AE (automatic exposure calculation) to set shooting conditions. Thereafter, in response to the release button being fully pressed, the CPU 31 captures the subject, and the image processing unit 20 generates captured image data. Then, the CPU 31 records the captured image data on the recording medium 34 and proceeds to S110. When a face area is detected in S102, it is preferable that the AF control unit 31a of the CPU 31 performs AF using the detected face area as a focus detection area.

  Step 108: In this case, the CPU 31 shifts to the first photographing mode suitable for photographing a child and photographs the subject. In the first shooting mode, the CPU 31 performs shooting in the same flow as the normal shooting mode in S107, but the CPU 31 is different in that it executes any one of the following controls (1) to (11).

  (1) The CPU 31 corrects the exposure time to be shorter than that in the normal shooting mode when setting the shooting conditions. By shortening the exposure time as described above, it is possible to suppress subject blurring caused by the movement of the child.

  For example, in the first shooting mode, the CPU 31 increases the imaging sensitivity as compared to the normal shooting mode, and decreases the aperture value to shorten the exposure time. Of course, the CPU 31 may shorten the exposure time by performing only one of the correction of the imaging sensitivity and the correction of the aperture value.

  (2) When setting the shooting conditions, the CPU 31 increases the depth of field by increasing the aperture value compared to the normal shooting mode. In a scene where a child is photographed, the child may move and it may be difficult to focus. However, by increasing the depth of field, the possibility of photographing failure can be reduced. Note that the settings of (1) and (2) above can be switched between each other by the operation of the operation unit 24 by the user.

  (3) The CPU 31 moves the photographing lens 11 to a position suitable for close-up photography. This is because, particularly in a scene where an infant is photographed, close-up photography is often performed with the camera close to a person.

  (4) The CPU 31 makes a setting to output the sound from the speaker 28 while causing the flash light emitting unit 29 to emit light immediately before the start of exposure. This is to prevent the shooting failure by directing the child's attention to the electronic camera during shooting.

  (5) The CPU 31 prohibits the flash light emission unit 29 from emitting light at the time of shooting, and performs a setting to turn off the sound output of the speaker 28. For example, when taking a picture of a sleeping child, the child may not be awakened by a flash or shooting sound at the time of shooting. Note that the settings of (4) and (5) above can be switched between each other by the operation of the operation unit 24 by the user.

  (6) The AF control unit 31 a of the CPU 31 executes AF only with the external light AF unit 30. That is, in the first shooting mode, the CPU 31 can be configured not to perform AF in the contrast detection method that takes time until focusing. Thereby, since AF is performed quickly, the possibility of missing a photo opportunity is reduced.

  (7) The CPU 31 estimates the facial expression of the main subject. Then, the CPU 31 enables photographing only when the facial expression of the main subject is in a predetermined state (for example, a state where the subject is not blinking or a smiling state). As a result, the user can obtain a captured image in which the expression of the main subject is in a desired state, and the possibility of shooting failure is reduced.

  Here, the detection of the facial expression is performed by the following means. For example, as disclosed in Japanese Patent Application Laid-Open No. 2004-272933, learning face images classified into classes such as a normal expression, a face with closed eyes, a face with open mouth, and a smile are prepared in advance, and the CPU 31 canonically. After performing discrimination by discriminant analysis, facial expressions may be discriminated based on evaluation values set for each class. Alternatively, standard patterns such as open eyes, closed eyes, open mouths, closed mouths may be prepared in advance, and the CPU 31 may match each pattern with the face area and estimate the facial expression from the combination of patterns. .

  (8) The CPU 31 executes continuous shooting of a plurality of frames by pressing the release button once. As a result, the user can select the best image from a plurality of photographed images, so that the possibility of photographing failure is reduced.

  (9) The CPU 31 performs recording with the microphone 27 during shooting. For example, the CPU 31 performs recording before and after capturing a captured image, and records the recorded image data in association with the captured image data. As a result, the situation of the photographed scene can be grasped by voice.

  (10) The CPU 31 instructs the image processing unit 20 to perform soft focus processing on the captured image. As a result, it is possible to obtain a preferred photographed image with the child being the main subject emphasized and soft.

  Here, the soft focus processing of the image processing unit 20 is performed by the following means. For example, the image processing unit 20 generates a soft focus image by performing filter processing using a filter that exhibits a predetermined soft focus effect, a low-pass filter, or the like on the captured image. The image processing unit 20 may generate a mask image from the captured image, and may generate a soft focus image by synthesizing the filtered image and the original image based on the mask image.

  (11) The CPU 31 instructs the image processing unit 20 to perform an auxiliary display composition process. Then, the image processing unit 20 synthesizes a character such as a cartoon or news on the shooting date with the shot image. Thereby, a photographed image with high entertainment property can be provided to the user.

  Step 109: In this case, the CPU 31 shifts to the second photographing mode suitable for female photographing and photographs the subject. In the second shooting mode, the CPU 31 performs shooting in substantially the same flow as in the normal shooting mode in S107, but is different in that the CPU 31 executes one of the following controls (1) to (4).

  (1) The CPU 31 instructs the image processing unit 20 to perform an exposure correction process for correcting the exposure of the captured image to the over side. This makes it possible to obtain a preferable captured image in which the skin of the main subject woman is expressed white.

  (2) The CPU 31 instructs the image processing unit 20 to perform soft focus processing on the captured image. As a result, it is possible to obtain a preferred photographed image with emphasis on the main female subject and softness. In addition, it is possible to obtain a preferable photographed image in which wrinkles and rough skin of a main subject woman are not noticeable. Note that the description of the soft focus process is the same as that in S107 described above, and thus a duplicate description is omitted.

  (3) The CPU 31 instructs the image processing unit 20 to correct the saturation parameter of the captured image. For example, the image processing unit 20 performs correction that lowers the saturation of the skin color region to express the skin white. Alternatively, the image processing unit 20 may perform correction to increase the saturation of the skin color area to reduce dullness of the skin color.

  (4) The CPU 31 instructs the image processing unit 20 to execute correction to make the person in the captured image look slim. As a result, it is possible to obtain a preferable photographed image that makes the main subject woman look thin.

  Here, the process of making the photographed person look slim is performed by the following means. For example, the image processing unit 20 adjusts the image size by trimming the image after slightly changing the aspect ratio of the captured image to a vertically long image. Alternatively, the image processing unit 20 may correct the outline of the face by image processing to make the cheek thinner as disclosed in Japanese Patent Application Laid-Open No. 2004-264893.

  Step 110: The CPU 31 determines whether or not generation of print setting data according to the DPOF (Digital Print Order Format) standard is set to ON. If the above setting is on (YES side), the process proceeds to S111. On the other hand, when the setting is off (NO side), the CPU 31 ends the photographing operation without generating print setting data.

  Step 111: The CPU 31 determines whether or not the main subject is an elderly person from the age estimated in the inference process of S104. Specifically, the CPU 31 determines that the main subject is an elderly person when the estimated age of the main subject is equal to or greater than a threshold (for example, 60 years or older). If the main subject is an elderly person (YES side), the process proceeds to S112. On the other hand, when the face area is not detected or when the main subject is not an elderly person (NO side), the process proceeds to S113.

  Step 112: The CPU 31 changes the print size to a size larger than the normal size and generates print setting data. For example, when the standard print size is L size, the CPU 31 generates print setting data with the print size set to 2L size. As a result, it is possible to print out a captured image that is easy to see for the elderly by the print setting data.

  Step 113: The CPU 31 generates print setting data in the normal print size. For example, when the standard print size is L size, the CPU 31 generates print setting data with the print size set to L size. This is the end of the description of the photographing operation of the electronic camera of the first embodiment.

(Description of Second Embodiment)
FIG. 3 is a flowchart for explaining an example of the photographing operation of the electronic camera in the second embodiment. In the second embodiment, the person's age and gender inference function and the face recognition function are set to ON and the registered person is recognized.

  Here, since the basic configuration of the electronic camera of the second embodiment is the same as that of the first embodiment, the duplicate description is omitted. Also, S201 and S202 in FIG. 3 correspond to S101 and S102 in the first embodiment, respectively, and thus redundant description is omitted.

  Step 203: The CPU 31 determines whether or not a face area is detected in the face detection process of S202. If a face area is detected (YES side), the process proceeds to S204. On the other hand, if the face area is not detected (NO side), the process proceeds to S208.

  Step 204: The face recognition unit 31c of the CPU 31 is registered as a recognition target from the face areas based on the feature point data acquired in the face detection process of S202 and the face recognition data of the second memory 32. A face recognition process for detecting a human face is executed.

  Here, in the face recognition process of S204, the CPU 31 corrects the threshold value of the determination criterion for the face recognition process for each registered person based on the characteristic data in the second memory 32. For example, for a registered person whose gender of the characteristic data is female, the CPU 31 corrects the threshold value of the determination criterion to be low and increases the possibility of determining a person in the face area as a registered person. This is because, in the case of women, the accuracy rate of face recognition may be lowered due to changes in makeup or hairstyle. For a registered person whose characteristic data indicates wearing glasses, the CPU 31 corrects the threshold value of the determination criterion to be low, and increases the possibility of determining a person in the face area as a registered person. This is because in the case of a person wearing spectacles, there is a possibility that the accuracy rate of face recognition may be lowered due to the presence or absence of spectacles or changes in spectacles.

  Step 205: The CPU 31 determines whether or not the face of the registered person has been detected in the face recognition process of S204. When the face of the registered person is detected (YES side), the process proceeds to S206. On the other hand, when the face of the registered person is not detected (NO side), the process proceeds to S208.

  Step 206: The inference unit 31d of the CPU 31 executes an inference process for estimating the sex, race, and age of the subject for the face area recognized as a registered person in S204. As an example, the CPU 31 sets a plurality of classes for each category of the subject's sex, race, and age, and obtains the likelihood for each class. Then, the CPU 31 estimates the sex, race, and age of the subject based on the class with the highest likelihood.

  Step 207: The CPU 31 determines whether or not the characteristic data in the second memory 32 matches the gender, race, and age estimated in S206. For example, the CPU 31 calculates an evaluation value indicating the degree of coincidence between the estimation result of S206 and the characteristic data for the sex, race, and age of the subject. If the evaluation value is greater than or equal to the threshold value, it is determined that the face area in S204 matches the registered person. If they match (YES side), the process proceeds to S209. On the other hand, if the two do not match (NO side), the process proceeds to S208.

  Step 208: In this case, the CPU 31 shoots the subject in the normal shooting mode. In the normal shooting mode, the CPU 31 performs AF and executes AE to set shooting conditions in response to half-pressing of the release button. Thereafter, in response to the release button being fully pressed, the CPU 31 captures the subject, and the image processing unit 20 generates captured image data. The CPU 31 records the captured image data on the recording medium 34. When a face area is detected in S202, it is preferable that the AF control unit 31a of the CPU 31 performs AF using the detected face area as a focus detection area.

  Step 209: In this case, the CPU 31 shifts to the registered person photographing mode and photographs the subject. In this registered person shooting mode, the CPU 31 performs shooting in substantially the same flow as in the normal shooting mode in S208, but the CPU 31 reads out the registered person's characteristic data detected in S204 from the second memory 32 and uses it as the characteristic data. Based on this, it is different in that one of the following controls (1) to (4) is executed.

  (1) The CPU 31 outputs a cry indicating a shooting signal from the speaker 28 during shooting. Here, the CPU 31 changes the above call based on the characteristic data of the registered person. Specifically, the CPU 31 selects audio data to be reproduced from the audio data in the second memory 32 based on the birthplace / language of the registered person of the characteristic data. For example, when the language in the characteristic data indicates English, the CPU 31 reproduces the English call voice data. Further, when a plurality of dialects are set for one language, the CPU 31 may reproduce the voice data of the dialect corresponding to the birthplace based on the birthplace of the registered person of the characteristic data. Thereby, it is possible to easily transmit the shooting signal to the registered person as the main subject.

  (2) The CPU 31 instructs the image processing unit 20 to display the shooting date. The image processing unit 20 combines the date acquired from the timer 31e with a predetermined position of the captured image. Here, the CPU 31 changes the display order of the date based on the characteristic data of the registered person. For example, when the language of the registered person in the characteristic data is Japanese, the CPU 31 describes the shooting dates in the order of year, month, and day. In addition, when the language of the registered person of the characteristic data is English, the CPU 31 describes the shooting dates in the order of day / month / year or month / day / year. Thereby, it is possible to provide a captured image in which the date and time are displayed in an easy-to-understand manner for the registered person as the main subject.

  (3) The CPU 31 adds the date and time calculated from the birthday to the captured image based on the date of birth of the registered person. For example, the CPU 31 instructs the image processing unit 20 to perform image processing for synthesizing characters indicating the age (for example, xx years old, age x months) with the photographed image. Or you may make it CPU31 record the data of the date calculated from a birthday on the data of a picked-up image. Thereby, the growth of the registered person as the main subject can be easily grasped, and a photographed image suitable for album editing or the like can be provided.

  (4) The CPU 31 instructs the image processing unit 20 to perform an auxiliary display composition process. Then, the image processing unit 20 synthesizes a character or the like designated for each registered person in the characteristic data with the captured image. Note that the image processing unit 20 can also change the combined position of the accompanying display for each registered person based on the characteristic data. Thereby, a photographed image with high entertainment property can be provided to the user.

  When combining characters such as the shooting date with the shot image in S209, the image processing unit 20 changes the font, color, size, and combining position of the characters for each registered person based on the character display item of the characteristic data To do. Above, description of imaging | photography operation | movement of the electronic camera of 2nd Embodiment is complete | finished.

  In the electronic camera of the second embodiment, face recognition processing is performed based on the feature amount detected from the face area, and the race, age, and gender estimated from the face area are collated with the characteristic data to determine the face. Check the validity of the recognition result. Therefore, the accuracy of the face recognition process can be further improved.

  In the second embodiment, the threshold value of the determination criterion for the face recognition process is corrected for each registered person based on information such as the gender of the characteristic data and the presence or absence of glasses. Therefore, the possibility that the registered person is not detected in the face recognition process can be reduced.

(Supplementary items of the embodiment)
(1) In the second embodiment, the example of executing the process of estimating the race, age, sex, etc. of the subject together with the face recognition process has been described. However, in the present invention, the process of estimating the race, etc. of the subject is omitted. You may do it.

  (2) The processing in the photographing mode of the registered person in the second embodiment is merely an example. For example, the shooting conditions at the time of shooting, image processing settings, settings for automatic image transmission by e-mail, and the like may be recorded in the second memory 32 as the characteristic data for each registered person. The CPU 31 may automatically set shooting conditions and image processing based on the characteristic data at the time of face recognition, or may automatically send shot image data to a predetermined address after shooting.

DESCRIPTION OF SYMBOLS 11 ... Shooting lens, 12 ... Lens drive part, 13 ... Aperture, 14 ... Aperture drive part, 15 ... Image sensor, 16 ... Analog processing part, 20 ... Image processing part, 27 ... Microphone, 28 ... Speaker, 29 ... Flash emission , 30 ... external light AF unit, 31 ... CPU, 31a ... AF control unit, 31b ... face detection unit, 31c ... face recognition unit, 31d ... inference unit, 31e ... timer, 32 ... second memory

Claims (6)

An image sensor for capturing a subject image;
And registration data indicating the characteristic amount of the face, are recorded correspondence relation between the characteristic data relating to only locations or language out, a memory for correlation between the sound indicated cues of the characteristic data and the photographing is recorded,
A face detecting unit that detects a face area of the subject image based on output of the imaging device, extracts a feature amount of the face of the subject image from the face region,
A recognition unit for recognizing the subject image based on the facial feature amount of the subject image extracted by the face detection unit and the registered data recorded in the memory ;
A speaker that outputs sound to the outside,
Based on the registered data of the object image is recognized by the prior Ki認 識部, a control unit for outputting a sound indicating the cues recorded the shot in the memory before recording from the speaker,
An electronic camera comprising: The electronic camera according to claim 1,
The control unit outputs a sound indicating the corresponding shooting signal from the speaker before shooting based on the characteristic data corresponding to the registration data of the subject image recognized by the recognition unit.
An electronic camera characterized by that. The electronic camera according to claim 2,
The control unit causes the speaker to output a sound indicating the shooting signal based on the birth place of the subject image recognized by the recognition unit from the speaker before shooting.
An electronic camera characterized by that. The electronic camera according to claim 3,
The control unit causes the speaker to output a sound indicating the shooting signal in a dialect corresponding to the birth place of the subject image recognized by the recognition unit before shooting.
An electronic camera characterized by that. An electronic camera according to any one of claims 1 to 4,
A timer to acquire the shooting date and time,
Based on the recognition result of the recognition unit, for the image for recording taken by the imaging device,
An image processing unit that executes image processing for combining display of shooting dates;
The control unit controls the image processing unit to change the display order of the shooting dates based on the characteristic data corresponding to the registration data of the subject image recognized by the recognition unit.
An electronic camera characterized by that. An electronic camera according to any one of claims 1 to 4,
An inference unit that performs an inference process on either the age or gender of the subject image based on the facial feature amount of the subject image extracted by the face detection unit;
The memory further records a correspondence relationship between the second characteristic data relating to any of gender, race, and age of the person and the registration data,
  The control unit is configured to determine the subject image and the registration data based on the inference result of any one of gender, race, and age of the subject image by the inference unit and the second characteristic data recorded in the memory. Determine whether or not
An electronic camera characterized by that.

Images