JP2017085204A - Registration control of meta-data - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an imaging device which reduce a lode of an image retrieval with meta-data by simplifying a registration control of the meta-data and registering only meta-data selected and designated by a user.SOLUTION: An imaging device comprises: photographing means of photographing an image imaged by an imaging optical system; image recognition means of detecting image information from the image; image recognition region display means of displaying the image information acquired by the image recognition means as image recognition region information; meta-data selection means of selecting the image recognition region information as a registration candidate of meta-data; and meta-data registration means of registering the meta-data selected by the meta-data selection means corresponding to the photographing image.SELECTED DRAWING: Figure 3

Description

  The present invention relates to an imaging apparatus and a control method thereof, and more particularly to a control method for registering image recognition information obtained from a captured image as metadata.

  2. Description of the Related Art Conventionally, there is a method in which a user manually inputs shooting information regarding a subject of image data shot by a digital camera as metadata.

  However, the input of shooting information has become more complicated for the user as the number of shot images and the number of subjects in each shot image data increase.

  In order to avoid complicated input operations, there is a method of recording text data obtained by character recognition from a photographed image as image metadata. Patent Document 1 proposes an imaging apparatus that registers take information written on a clapperboard in image metadata. Patent Document 2 proposes a method in which a face detected from an image is displayed with a frame, and a keyword such as a person's name is recorded as image metadata in association with the face frame selected by touch. .

JP 2005-39354 A JP 2010-252011

  However, in the prior art disclosed in Patent Document 1 described above, all character string information detected by image recognition of a captured image is recorded as metadata, so that unnecessary information is also registered as metadata. Data is not given properly. When metadata unintended by the user is registered, there is a problem that extra metadata information hits when searching for an image using metadata, making it difficult to search. In the prior art disclosed in Patent Document 2 described above, the subject information to be registered in the metadata can be selected, but the subject information needs to be manually input, which is a complicated operation.

  The present invention has been made in order to solve the above-mentioned problems, simplify the registration control of metadata, and register only the metadata selected and specified by the user, thereby reducing the load of image search using metadata. The purpose is to reduce.

In order to achieve the above object, an imaging apparatus according to the present invention includes:
Image capturing means for capturing an image formed by the imaging optical system, image recognition means for detecting image information from the image, and image recognition area for displaying image information obtained from the image recognition means as image recognition area information Display means; metadata selection means for selecting the image recognition area information as a metadata registration candidate; and metadata registration means for registering the metadata selected by the metadata selection means in association with the photographed image. It is characterized by having.

  According to the imaging device of the present invention, a cursor frame or the like is displayed on the display screen for the person and character string information detected from the captured image, and the information selected by the user is registered in the metadata. Of the information detected from the recognition, only the information specified by the user is registered as metadata, thereby providing a function of reducing the load of image search using metadata.

System configuration diagram of digital camera of embodiment External view of digital camera of embodiment A flowchart showing a sequence of “selective registration mode” in metadata registration according to the first embodiment. 9 is a flowchart showing a sequence of “selection registration notification mode” in metadata registration according to the second embodiment. 10 is a flowchart showing a sequence of “image area determination mode” in metadata registration according to the third embodiment. A flowchart showing a sequence of “person discrimination mode” in metadata registration according to the fourth embodiment. Explanatory drawing of "selection registration mode" in metadata registration of Example 1 Explanatory drawing of "person discrimination mode" in the metadata registration of Example 4

  Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[Example 1]
FIG. 1 shows a system configuration diagram of a digital camera to which the present invention can be applied.

  The subject image passes through the optical lens (10) and is then formed on the image sensor (11). The imaging control circuit (12) controls the drive of the image sensor (11), converts the image signal from the image sensor (11) from analog to digital, and sends it to the system control circuit (20). The image sensor (11) is composed of a CCD or a CMOS, and converts a subject image into an image signal and outputs it.

  A character recognition circuit (13) obtains character string information by cutting out a character region from a captured image and performing character recognition. The face recognition circuit determines the face area of a person and performs face authentication of the person by known face recognition control. Under the control of the system control circuit (20), the memory control circuit (21) is a volatile memory (25), an image sensor (11), an image processing circuit (22), a recording memory (23), a display unit ( Control data transmission / reception in 24). The system control circuit (20) is a circuit for controlling the entire camera, and also serves as an arithmetic circuit for performing various arithmetic operations.

  The volatile memory (25) is a high-speed accessible memory such as SDRAM. The nonvolatile memory (26) is a memory for recording constants, variables, and programs for the operation of the system control circuit (20). The recording memory (23) is a memory that can be detached from the digital camera. If the memory capacity is insufficient, it can be replaced. The operation of the system control circuit (20) is determined by an instruction signal from the release operation button (15), MENU operation button (16), selection operation button (17), selection decision button (18), and metadata registration button (19). Is done. The MENU operation button (16) displays a screen for setting various cameras and is used for setting each mode.

  FIG. 2 shows an external view of the digital camera.

  A display unit (24) in FIG. 1 is provided on the back of the digital camera body (201). Further, a release operation button (15), a MENU operation button (16), a selection operation button (17), a selection determination button (18), and a metadata registration button (19) in FIG. 1 are provided.

  There are two image recognition modes: a metadata selection registration mode and a metadata automatic registration mode. The metadata selection / registration mode is a mode in which information selected by the user among the person and character string information detected from the photographed image is registered in the metadata. On the other hand, the metadata automatic registration mode is a mode in which all of the person and character string information detected from the captured image are automatically registered in the metadata. The metadata selection registration mode has four modes.

  The first mode displays a frame etc. on the display screen for the person and character string information detected from the photographed image, and registers the information selected by the user in the metadata in “selection registration mode” is there. The second mode is a “selection registration notification mode” that calculates the total number of persons and character string information detected from the captured image and determines whether or not to perform selective registration of metadata based on the calculated total number. is there. The third mode is an “image region discrimination mode” in which the size of an image region for performing image recognition is set in advance, and detection is performed by narrowing down a person having a predetermined image recognition region size and character string information from a captured image. . The fourth mode is a “person discrimination mode” in which a face having a predetermined face recognition area size is identified as a person using face recognition area information, photographing lens focal length information, and subject distance information from the photographed image. It is.

  FIG. 3 illustrates a shooting and metadata registration sequence according to the present embodiment with reference to a flowchart. This flowchart is a “selection registration mode” sequence in which a frame or the like is displayed on the display screen for the person and character string information detected from the photographed image, and the information selected by the user is registered in the metadata.

  When shooting is started, a sequence according to the flowchart is started. When the imaging process is performed by the imaging control circuit (12) (S301), the process proceeds to the next (S302), where the captured image is stored and stored as image data in the recording memory (23). Next, the process proceeds to (S303), and the image stored in the recording memory (23) is displayed on the display unit (24). Next, the process proceeds to (S304), and the face recognition circuit (14) detects the face from the display image. If the face is recognized, the process proceeds to (S305), and a frame such as a cursor is displayed in the face area of the person. Next, the process proceeds to (S306), and whether or not there is an operation for selecting a metadata registration candidate using the selection operation button (17) and the selection determination button (18) for a frame area such as a cursor displayed on the face area of the person. Determine. When the selection operation of the metadata registration candidate is completed by the selection decision button (18), the selection data is stored in the nonvolatile memory (26), and the process proceeds to the next (S307).

  On the other hand, if the registration candidate selection operation is a cancel operation (not shown) or the metadata registration candidate selection operation is not completed for a predetermined period in (S306), the series of sequences is terminated without registering metadata.

  On the other hand, if the face cannot be recognized in (S304), the process proceeds to (S307), and character string information is detected from the display image in the character recognition circuit (13). When the character string information is detected, the process proceeds to (S308), and a frame such as a cursor is displayed in the character recognition area. Next, the process proceeds to (S309), and the text data acquired from the detected character string information is displayed in a frame such as a cursor. Next, the process proceeds to (S310), and whether or not there is an operation for selecting a metadata registration candidate using the selection operation button (17) and the selection determination button (18) for a frame area such as a cursor displayed in the character recognition area. judge. When the selection operation of the metadata registration candidate is completed by the selection decision button (18), the process proceeds to the next (S311).

  On the other hand, if the registration candidate selection operation is a cancel operation (not shown) or the metadata registration candidate selection operation is not completed for a predetermined period in (S310), the series of sequences is terminated without performing metadata registration.

  On the other hand, if character recognition could not be performed in (S307), the process proceeds to (S311), and face detection by image recognition performed in (S306) and (S310) and determination of metadata candidate selection from character detection I do. If one or more metadata candidates are selected, the process proceeds to the next (S312), and the operation of the metadata registration button (19) is determined. When the operation of the metadata registration button (19) is performed, the process proceeds to (S313), the selection data of the metadata registration candidates selected in (S306) and (S310) is read from the nonvolatile memory (26), and the display image is displayed. Register in association with metadata.

  On the other hand, in (S311), when face detection from the display image in (S306) and (S310) and selection / detection of metadata candidates by character recognition are not performed, the series of sequences is completed without registering metadata. . On the other hand, if the metadata registration button (19) is not operated for a predetermined period in (S312), or if the metadata registration cancellation operation (not shown) is performed, the series of sequences is terminated without registering metadata. To do.

  FIG. 7 shows a display example of the image recognition information displayed in the present embodiment. FIG. 7A shows a state in which a photographed image is displayed. FIG. 7B shows an example of a state in which a cursor frame is displayed in the area of the person and character string information detected from the photographed image of FIG.

  (701) indicates the frame display of the human face area, and (702) and (703) indicate the cursor frame display of the character string information area. Further, an example is shown in which text data acquired from the character string information is displayed in the display of the cursor frame of the character string information indicated by (702) and (703). FIG. 7 (c) shows a state in which the cursor frame of (703) is deselected from the detected character string information of FIG. 7 (b), the cursor frame is erased and excluded from the metadata registration candidates.

  Information detected from image recognition by displaying a cursor frame etc. on the display screen for the person and character string information detected from the captured image as described above, and registering the information selected by the user in the metadata Among them, it is possible to provide a function of registering only information designated by the user as metadata.

[Example 2]
FIG. 4 illustrates a shooting and metadata registration sequence according to this embodiment with reference to a flowchart. This flowchart is a “selection registration notification mode” sequence in which the total number of persons and character string information detected from a captured image is calculated, and it is determined whether or not to perform selective registration of metadata based on the number.

  When shooting is started, a sequence according to the flowchart is started. When the imaging process is performed by the imaging control circuit (12) (S401), the process proceeds to the next (S402), the captured image is stored, and is stored in the recording memory (23) as image data. Next, the process proceeds to (S403), and the image stored in the recording memory (23) is displayed on the display unit (24). Next, proceed to (S404), face detection is performed from the display image in the face recognition circuit (14), and if face recognition is possible, proceed to (S405) and display a frame such as a cursor on the face area of the person. Go to (S406).

  On the other hand, if the face cannot be recognized in (S404), the process proceeds to (S406), and character string information is detected from the display image in the character recognition circuit (13). If the character string information can be detected, the process proceeds to (S407), and a frame such as a cursor is displayed in the character recognition area. Next, the process proceeds to (S408), the text data acquired from the detected character string information is displayed in the display of a frame such as a cursor, and the process proceeds to the next (S409).

  On the other hand, if character recognition could not be performed in (S406), the process proceeds to (S409), and face detection by image recognition performed in (S404) and (S406) and confirmation of metadata candidate detection from character detection I do. If metadata candidates are detected, proceed to the next (S410), calculate the total number of face areas and character recognition areas detected in (S405) and (S408), and calculate the total number of detection areas. Comparison with the threshold (Ath) is performed. If “Ath ≦ A” that is equal to or greater than the threshold value, the process proceeds to the next (S411), and a notification display indicating that the mode is shifted to a mode for selecting a metadata registration candidate is performed.

  Next, proceed to (S412), select operation buttons (17) and selection decision buttons (18) for the frame area such as the cursor displayed in the face area and character recognition area of the person displayed in (S405) and (S408) It is determined whether or not there is an operation for selecting a metadata registration candidate using. When the selection operation of the metadata registration candidate is completed by the selection decision button (18), the selection data is stored in the nonvolatile memory (26), and the process proceeds to the next (S413), and the operation of the metadata registration button (19) is performed. Make a decision. When the operation of the metadata registration button (19) is performed, the process proceeds to (S414), the selection data of the metadata registration candidate selected in (S412) is read from the nonvolatile memory (26), and the display image metadata is supported. Then register.

  On the other hand, when face detection and image recognition detection by character recognition are not performed from the display image in (S409) and (S406) in (S409), the series of sequences is terminated without registering metadata. On the other hand, in (S410), comparison with the threshold value (Ath) of the total number of detection areas is performed, and if `` Ath ≧ A '' below the threshold value, proceed to (S414) and be displayed in (S405) and (S408) The character string information of the person's face area and character recognition area is automatically registered in association with the metadata of the display image. On the other hand, in (S412), if the registration candidate selection operation is a cancel operation (not shown) or the metadata registration candidate selection operation is not completed for a predetermined period, the series of sequences is terminated without performing metadata registration.

  On the other hand, if the metadata registration button (19) is not operated for a predetermined period in (S413), or if the metadata registration cancellation operation (not shown) is performed, the series of sequences is terminated without registering metadata. To do.

  As described above, a frame or the like is displayed on the display screen for the person and character string information detected from the captured image, and a predetermined threshold is set for the total number of detected areas. Only when the number of areas equal to or greater than a predetermined threshold is detected, the process proceeds to a metadata registration candidate selection operation. When the number of areas is equal to or less than the threshold, all persons and character string information detected from the captured image are automatically registered in the metadata. Thus, it is possible to provide a function for reducing the load of metadata registration.

[Example 3]
FIG. 5 illustrates a shooting and metadata registration sequence according to the present embodiment with reference to a flowchart. This flowchart is a sequence of “image region discrimination mode” in which the size of an image region for performing image recognition is set in advance from a captured image, and only a person having a predetermined image recognition region size and character string information are detected.

  When shooting is started, a sequence according to the flowchart is started. When the imaging process is performed by the imaging control circuit (12) (S501), the process proceeds to the next (S502), where the captured image is stored and stored as image data in the recording memory (23). Next, the process proceeds to (S503), and the image stored in the recording memory (23) is displayed on the display unit (24). Next, proceed to (S504), perform face detection from the display image in the face recognition circuit (14), and if face recognition is possible, proceed to (S505), and set the area size of the face area detected in (S504). It is calculated and compared with the face area size threshold values (Kth1, Kth2).

  If “Kth1 ≦ K ≦ Kth2” within the threshold, the process proceeds to the next (S506), and a frame such as a cursor is displayed on the face area of the person. Next, proceed to (S507), and whether or not there is a metadata registration candidate selection operation using the selection operation button (17) and selection decision button (18) for the frame area such as the cursor displayed on the face area of the person Determine. When the selection operation of the metadata registration candidate is completed by the selection decision button (18), the selection data is stored in the nonvolatile memory (26), and the process proceeds to the next (S507).

  On the other hand, if the registration candidate selection operation is a cancel operation (not shown) or the metadata registration candidate selection operation is not completed for a predetermined period in (S507), the series of sequences is terminated without performing metadata registration. On the other hand, if the face area size is other than the threshold value “Kth1 ≦ K ≦ Kth2” in (S505), the process proceeds to (S508). On the other hand, if the face cannot be recognized in (S504), the process proceeds to (S508), and character string information is detected from the display image in the character recognition circuit (13).

  If the character string information can be detected, the process proceeds to (S509), where the area size of the character area detected in (S508) is calculated, and compared with the threshold values (Mth1, Mth2) of the character area size. If “Mth1 ≦ M ≦ Mth2” within the threshold, the process proceeds to the next (S510), and a frame such as a cursor is displayed in the character recognition area of the designated area size. Next, proceeding to (S511), the text data acquired from the detected character string information is displayed in the display of a frame such as a cursor. Next, proceed to (S512) and check whether there is a metadata registration candidate selection operation using the selection operation button (17) and selection decision button (18) for the frame area such as the cursor displayed in the character recognition area. judge. When the selection operation of the metadata registration candidate is completed by the selection decision button (18), the process proceeds to the next (S513).

On the other hand, if the registration candidate selection operation is a cancel operation (not shown) or the metadata registration candidate selection operation is not completed for a predetermined period in (S512), the series of sequences is terminated without registering metadata. On the other hand, if the character area size is other than the threshold value “Mth1 ≦ M ≦ Mth2” in (S509) (S513
Go to).

  On the other hand, if character recognition could not be performed in (S508), proceed to (S513), face detection by image recognition performed in (S507) and (S512), and determination of whether metadata candidates are selected from character detection I do. When one or more metadata candidates are selected, the process proceeds to the next (S513), and the operation of the metadata registration button (19) is determined. When the operation of the metadata registration button (19) is performed, the process proceeds to (S514), the selection data of the metadata registration candidate selected in (S507) and (S512) is read from the nonvolatile memory (26), and the display image is displayed. Register in association with metadata.

  On the other hand, in (S513), if face detection from the display image in (S507) and (S512) and selection of metadata candidates by character recognition are not detected, the series of sequences is terminated without registering metadata. . On the other hand, if the metadata registration button (19) is not operated for a predetermined period in (S514), or if the metadata registration cancellation operation (not shown) is performed, the series of sequences is terminated without registering the metadata. To do.

  As described above, by setting a threshold for the detection area size for detecting a face or character from a captured image, the information to be detected as metadata registration candidates is narrowed down, and the metadata registration load is reduced by performing metadata selective registration. The function to reduce can be provided.

[Example 4]
FIG. 6 illustrates a shooting and metadata registration sequence according to the present embodiment with reference to a flowchart. This flowchart is a “person discrimination mode” for discriminating a face of a predetermined face recognition area size from a person using face recognition area information detected from a photographed image, focal length information of a photographing lens, and subject distance information. This is the sequence.

  When shooting is started, a sequence according to the flowchart is started. When imaging processing is performed in the imaging control circuit (12) (S601), the process proceeds to the next (S602), and the face recognition circuit (14) detects the face within the shooting angle of view and calculates the area size of the face area. To do. Next, the process proceeds to (S603), and the subject distance information to the face area calculated in (S602) is acquired. As a method for acquiring subject distance information, there is a known method using reflection of near-infrared light. Next, proceeding to (S604), the focal length information of the optical lens (10) is acquired.

  Next, proceeding to (S605), the imaging control circuit (12) performs imaging processing, and performs captured image storage processing. The storage process associates the face recognition information detected in (S602), the subject distance information acquired in (S603), and the lens focal distance information acquired in (S604) into metadata, which is recorded together with the image data. Save to memory (23). Next, the process proceeds to (S606), and the image stored in the recording memory (23) is displayed on the display unit (24).

  Next, proceed to (S607), calculate the actual size of the face recognition area size using the area size of the face area associated with the image displayed in (S606), the subject distance information, and the focal length information. The person's face is discriminated by comparison with size thresholds (Hth1, Hth2). If “Hth1 ≦ H ≦ Hth2” within the threshold, the process proceeds to the next (S608), and a frame such as a cursor is displayed in the face recognition area determined to be a person. Next, proceed to (S609), and whether or not there is a metadata registration candidate selection operation using the selection operation button (17) and selection decision button (18) for the frame area such as the cursor displayed on the face area of the person Determine. When the selection operation of the metadata registration candidate is completed by the selection determination button (18), the selection data is stored in the nonvolatile memory (26), and the process proceeds to the next (S610).

  On the other hand, if the registration candidate selection operation is a cancel operation (not shown) or the metadata registration candidate selection operation is not completed for a predetermined period in (S610), the series of sequences is terminated without registering metadata.

  On the other hand, if the face recognition area size is other than the threshold value “Hth1 ≦ H ≦ Hth2” in (S607), the process proceeds to (S610), and character string information is detected from the display image in the character recognition circuit (13). . If the character string information can be detected, the process proceeds to (S611), and a frame such as a cursor is displayed in the character recognition area. Next, proceeding to (S612), the text data acquired from the detected character string information is displayed in the display of a frame such as a cursor.

  Next, proceed to (S613) and check whether there is an operation for selecting a metadata registration candidate using the selection operation button (17) and the selection decision button (18) for the frame area such as the cursor displayed in the character recognition area. judge. When the selection operation of the metadata registration candidate is completed by the selection decision button (18), the process proceeds to the next (S614).

  On the other hand, if the registration candidate selection operation is a cancel operation (not shown) or the metadata registration candidate selection operation is not completed for a predetermined period in (S614), the series of sequences is terminated without registering metadata.

  On the other hand, if character recognition could not be performed in (S610), the process proceeds to (S614), and face detection by image recognition performed in (S609) and (S613) and determination of metadata candidate selection from character detection I do. When one or more metadata candidates are selected, the process proceeds to the next (S615), and the operation of the metadata registration button (19) is determined. When the operation of the metadata registration button (19) is performed, the process proceeds to (S616), the selection data of the metadata registration candidates selected in (S609) and (S613) is read from the nonvolatile memory (26), and the display image is displayed. Register in association with metadata.

  On the other hand, in (S614), if face detection from the display image in (S609) and (S613) and selection / detection of metadata candidates by character recognition are not performed, the series of sequences is terminated without registering metadata. . On the other hand, if the face recognition area size in (S607) is other than the threshold value “Hth1 ≦ H ≦ Hth2” in (S614), and if image recognition detection by character recognition is not performed from the display image in (S610), metadata A series of sequences is completed without registering. On the other hand, if the metadata registration button (19) is not operated for a predetermined period in (S615), or if the metadata registration cancellation operation (not shown) is performed, the series of sequences is terminated without registering the metadata. To do.

FIG. 8 shows a display example of the image recognition information displayed in the present embodiment. FIG. 8 is a diagram showing a state where there are three face detection candidates in the photographed image.
(801) specifies a face area of a person and shows a frame display in the face area. (802) shows a state in which a poster with a person drawn on the wall of the room is pasted, but the face of the poster is not detected as a person's face, and a frame is not displayed in the face area. (803) shows a state where the doll is placed on the chair, but the face of the doll is not detected as a human face, and a frame is not displayed in the area of the face of the doll.

  As described above, the face detection information detected from the captured image is calculated by calculating the actual size of the face from the area size of the face area associated with the image, subject distance information, and focal length information, and specifying the face of the person. A function for reducing the load of data registration can be provided.

  In the embodiment described above, the person and the character string information detected from the captured image are displayed by displaying a frame or the like on the display screen. However, the operation is not limited to this embodiment. For example, what is detected from the photographed image may be other than person or character string information. Moreover, it is good also as a structure which displays the information detected from the picked-up image by a list etc., and performs selection operation. In addition, as a method for selecting as metadata registration candidates, a method for selecting and excluding unnecessary information from a person or character string information detected from an image has been described. However, only necessary information may be selected. Also, regarding the selection operation to be selected as the metadata registration candidate, an example in which the selection operation is specified using the button operation has been shown, but the configuration may be such that it is specified by an operation using the touch panel.

  In addition, a frame or the like is displayed on the display screen for the person and character string information detected from the captured image, and a predetermined threshold is set for the total number of detected areas. Only when the number of areas equal to or greater than a predetermined threshold is detected, the process proceeds to a metadata registration candidate selection operation. When the number of areas is equal to or less than the threshold, all persons and character string information detected from the captured image are automatically registered in the metadata. Thus, it is possible to provide a function for reducing the load of metadata registration. Also, by setting a threshold for the detection area size for detecting faces and characters from captured images, the information to be detected as metadata registration candidates is narrowed down, and metadata registration load is reduced by selecting and registering metadata. Function can be provided.

  In addition, for face detection information detected from a captured image, metadata registration is performed by calculating the actual size of the face from the area size of the face area associated with the image, subject distance information, and focal length information, and specifying the person's face It is possible to provide a function for reducing the load on the device.

(Other examples)
The present invention supplies a program that realizes one or more functions of the above-described embodiments to a system or apparatus via a network or a storage medium, and one or more processors in a computer of the system or apparatus read and execute the program This process can be realized. It can also be realized by a circuit (for example, ASIC) that realizes one or more functions.

10 optical lens, 11 image sensor, 12 imaging control circuit, 13 character recognition circuit,
14 face recognition circuit, 15 release operation buttons, 16 MENU operation buttons,
17 direction key operation buttons, 18 selection operation buttons, 19 metadata registration buttons,
20 System control circuit, 21 Memory control circuit, 22 Image processing circuit, 23 Recording memory,
24 display, 25 volatile memory, 26 non-volatile memory

Claims (5)

Photographing means (11) for photographing an image formed by the imaging optical system;
Image recognition means (20) for detecting image information from the image;
Image recognition area display means (24) for displaying the image information obtained from the image recognition means (20) as image recognition area information;
Metadata selection means (17) (18) for selecting the image recognition area information as metadata registration candidates;
Metadata registration means (19) for registering the metadata selected by the metadata selection means in association with the captured image;
An imaging device comprising: The image recognition means (20) is a face recognition means (14) for recognizing a human face from the image and detecting face area information;
Recognizing a character part from the photographed image, and having character recognition means (13) for acquiring character string information,
The image recognition area display means (24) displays face area information and character string information obtained by the face recognition means (14) and character recognition means (13) as image recognition area information. Item 2. The imaging device according to Item 1. The number of image recognition area information displayed by the image recognition area display means (24) is calculated, and when the image recognition area information is a predetermined number or more, a notification means (24) for displaying a notification is provided.
If a notification display is generated by the notification means (24), the process proceeds to the metadata selection means (17) (18). If a notification display does not occur, the process automatically proceeds to the metadata registration means (19). The imaging apparatus according to claim 1, wherein: The image recognition area display means (24) includes image recognition area size determination means (20) for determining the area size of the image recognition area information,
The image recognition area display means (24) uses the image recognition area size determination means (20), and does not display the image recognition area information as image recognition area information when the image recognition area information is a predetermined area size or less. The imaging device according to claim 1. Focal length information acquisition means (20) for acquiring focal length information from the imaging optical system lens;
Subject distance information acquisition means (11) for acquiring subject distance information is provided,
The imaging apparatus according to claim 1, wherein the face recognition means (14) detects area information of a person's face using the focal length information and the subject distance information.