JP2005323227A - Device, method, and program for picking up image - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To impart identification information used to classify the image without any trouble. <P>SOLUTION: Reference image data Ri obtained by photographing a person are stored in a reference image storage part 6 together with the imparted identification information representing the person. When an imaging part 2 acquires image data S0 through photography, a decision part 8 calculates the similarity between a face image included in an image that the image data S0 represent and the reference image Ri. Information on a reference image RTi whose similarity becomes equal to or above a threshold is inputted to an image recording part 10, which imparts the identification information imparted to the reference image data RTi representing the reference image RTi to the image data S0 and records them to media. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an image capturing device such as a digital camera that acquires an image by photographing, an image capturing method, and a program for causing a computer to execute the image capturing method.

  In a digital camera, an image acquired by shooting is recorded as digital image data in a recording medium such as an internal memory or an IC card provided in the digital camera, and an image acquired by shooting is recorded based on the recorded digital image data. Can be reproduced on a printer or monitor.

On the other hand, various methods for identifying a face have been proposed (see Patent Documents 1 and 2). By using such a method, face images of a plurality of persons are registered in advance, and by comparing the acquired face image with the registered face image, which person is included in the acquired image Can be identified.
Japanese Patent Laid-Open No. 7-23012 Japanese Patent Laid-Open No. 11-296664

  By the way, when shooting using a digital camera, the shooting target is often biased toward a specific subject. For example, in a family with children, there are particularly many children images and family images. In such a case, it is convenient to classify and organize the images by adding identification information for identifying the photographing target to the images. Here, a tag is attached to the image, and a comment or the like can be described in the tag. For this reason, it is possible to classify and organize images according to the shooting target by referring to the tag by describing identification information for identifying the shooting target as a comment using a personal computer.

  However, the operation of displaying the images one by one and describing the identification information as a comment is very cumbersome and time-consuming.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to provide identification information used when classifying an image to an image without taking time and effort.

An image capturing apparatus according to the present invention includes a reference image storage unit that stores a plurality of reference images each including a plurality of types of subjects in association with identification information for identifying subjects included in each of the plurality of reference images;
Imaging means for acquiring an image representing a subject by photographing;
A determination means for calculating a similarity between each of the acquired images acquired by the photographing and the plurality of reference images, and determining whether the similarity is equal to or higher than a predetermined threshold;
When the determination is affirmative, the information processing apparatus includes an information adding unit that adds the identification information associated with the reference image determined to be similar to the acquired image to the acquired image. is there.

  In the image pickup apparatus according to the present invention, the subject may be a person, and the identification information may be information for identifying the person.

An image capturing method according to the present invention acquires an image representing a subject by shooting,
Reference image storage means for storing an acquired image acquired by photographing and a plurality of reference images each including a plurality of types of subjects in association with identification information for identifying subjects included in each of the plurality of reference images. Calculating similarity with each of the stored reference images;
Determine whether the similarity is greater than or equal to a predetermined threshold,
When the determination is affirmative, the identification information associated with the reference image determined to be similar to the acquired image is added to the acquired image.

  The image capturing method according to the present invention may be provided as a program for causing a computer to execute the image capturing method.

  According to the present invention, when an image representing a subject is acquired by shooting, the similarity between the acquired image acquired by shooting and each of the plurality of reference images is calculated, and the calculated similarity is a predetermined threshold. It is determined whether or not the value is greater than or equal to the value. When this determination is affirmed, identification information associated with a reference image determined to be similar to the acquired image is given to the acquired image. For this reason, by photographing a desired subject in advance and obtaining a reference image, and storing the reference image in association with identification information for identifying the subject, the subject included in the stored reference image is photographed. As long as this is performed, identification information is given. Therefore, it is not necessary to perform an operation of assigning identification information to each image, and as a result, it is possible to perform an operation of classifying acquired images for each subject without trouble.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a schematic block diagram showing a configuration of a digital camera according to an embodiment of an image pickup apparatus of the present invention. As shown in FIG. 1, the digital camera 1 according to the present embodiment includes an imaging unit 2, an image storage unit 4, a reference image storage unit 6, a determination unit 8, an image recording unit 10, a display unit 12, The input unit 14 is connected to the bus 16.

  The imaging unit 2 includes an optical system, an imaging device such as a CCD, a shutter, an A / D conversion unit, and the like, and digital image data representing a subject by photographing the subject when the user of the digital camera 1 drives the shutter. Get S0.

  The image storage unit 4 serves as a work area when performing a determination process described later.

  The reference image storage unit 6 stores a plurality of reference image data Ri (i = 1 to n) representing a reference image. In the present embodiment, the reference image data Ri is acquired by the user photographing the faces of his / her family and relatives. At this time, in the reference image data Ri, identification information for identifying the family and relatives is described in the tag. For example, when the faces of the user's father, mother, wife, eldest son and second son are photographed and four reference image data R1 to R4 are acquired, the father, Described as mother, eldest son and second son.

  When the reference image storage unit 6 stores the reference image data Ri, the face contour is derived from the reference image represented by the reference image data Ri (hereinafter, the reference image Ri is also used as with the reference image data). Is detected and normalized so that the size of the region surrounded by the outline becomes a predetermined size, so that the size of the face included in all the reference images Ri and the size of the reference images Ri themselves are the same.

  The determination unit 8 calculates the similarity between the image represented by the image data S0 acquired by the imaging unit 2 (hereinafter, the reference image S0 is used similarly to the image data for the image) and the reference image Ri. It is determined whether or not the threshold value is Th1 or more. Specifically, the determination is performed as follows.

  First, the determination unit 8 detects a face outline from the image S0. The face contour is detected by detecting, for example, an oval skin color region included in the image and detecting the contour of the detected skin color region. Then, the detected face contour is normalized so as to have the predetermined size. If the image S0 includes a plurality of faces, the outlines of all the faces are normalized. Then, a face image having the same size as the reference image Ri (Fj, j = 1 to m (m is the number of faces included in the image S0)) is extracted from the normalized image S0, and the extracted face image Fj is extracted. Obtain the autocorrelation coefficient.

Here, the autocorrelation coefficient is calculated by the following equation (1).

In equation (1), P (u, v) is the image value of the face image Fj, Q and R are the size of the face image Fj, and (ξ, η) is the relative movement amount of the image. Here, when calculated in the range of −T ≦ ξ ≦ T and −T ≦ η ≦ T, (2T + 1) ² autocorrelation coefficients are obtained. (2T + 1) ^Two autocorrelation coefficients are arranged as a feature vector of the face image Fj.

  Further, the determination unit 8 obtains the feature vectors of all the reference images Ri in the same manner as described above. Note that the feature vector of the reference image Ri may be calculated in advance and stored in the reference image storage unit 6.

Then, the determination unit 8 calculates the similarity D between the feature vectors of all the reference images Ri and the feature vectors of the face image Fj by the following equation (2). In Equation (2), σ _k (ξ, η) is a feature vector of the kth reference image Rk.

  The determination unit 8 calculates the similarity D between all the face images Fj and the reference image Ri while sequentially changing the reference image Ri, and determines whether or not the similarity D is equal to or greater than the threshold value Th1. Then, the determination unit 8 outputs information representing the reference image RTi that is equal to or greater than the threshold Th1 to the image recording unit 10.

  Based on the information indicating the type of the reference image RTi input from the determination unit 8, the image recording unit 10 obtains the face image Fj using the identification information described in the tag of the reference image data RTi indicating the reference image RTi. After being described in the tag of the image data S0, the image data S0 is recorded on the medium.

  The display unit 12 includes a liquid crystal monitor that displays image data S0 and various commands.

  The input unit 14 includes a cross key, a dial, and various buttons, and can be used to set a mode (for example, a display mode, a shooting mode, and the like) of the imaging apparatus 1 and perform various inputs when operated by a user.

  Next, processing performed in the present embodiment will be described. First, registration of the reference image Ri will be described. First, the user switches the digital camera 1 according to the present embodiment to the shooting mode, and takes a picture of a person to be registered as the reference image Ri. After shooting, the digital camera 1 is switched to the display mode, and an image to be registered as the reference image Ri is displayed on the display unit 12. Then, by operating the input unit 14, a list of representative identification information is displayed on the display unit 12. Representative identification information includes, for example, grandfather, grandmother, father, mother, husband, wife, eldest son, second son, third son, eldest daughter, second daughter, third girl, and the like. When appropriate identification information is selected from the list and the input unit 14 is operated, the selected identification information is described in a tag of image data representing an image being displayed, and the image data is used as reference image data Ri as a reference image. The reference image Ri is registered by being stored in the storage unit 6. Note that identification information may be directly input instead of selection from the list. As the identification information, the name of a person photographed as a subject may be used.

  Next, processing performed at the time of shooting will be described. FIG. 2 is a flowchart showing processing performed at the time of photographing with the digital camera 1 according to the present embodiment. Here, it is assumed that the image S0 includes only one face.

  The process starts when the user drives the shutter of the digital camera 1 to shoot a subject. First, the imaging unit 2 acquires the image data S0 (step S1). Next, the determination unit 8 reads the reference image data Ri from the reference image storage unit 6 in order from i = 1 (step S2), and calculates the similarity D between the face image Fj extracted from the image S0 and the reference image data Ri (step S2). S3).

  Then, it is determined whether or not the calculated similarity D is equal to or greater than the threshold value Th1 (step S4). When step S4 is affirmed, information indicating the reference image RTi for which the similarity D has been calculated is stored in the image recording unit 10. (Step S5). The image recording unit 10 adds the identification information to the image data S0 by describing the identification information described in the tag of the reference image data RTi representing the reference image RTi in the tag of the image data S0 obtained from the face image Fj ( In step S6), the image data S0 to which the identification information is added is recorded on the medium (step S7), and the process ends.

  If step S4 is negative, the determination unit 8 determines whether or not there is the next reference image data Ri (step S8), and if step S8 is positive, the process returns to step S2. If step S9 is negative, the process ends.

  As described above, in this embodiment, when the similarity D between the face image Fj and the reference image Ri included in the image S0 acquired by photographing is equal to or greater than the threshold Th1, the reference image Ri is given to the image S0. Since the identification information is added, the identification information is added to the image S0 acquired by the shooting as long as the shooting is performed. Therefore, it is not necessary to perform the operation of assigning identification information to each image, and as a result, it is possible to perform the operation of classifying the image S0 for each subject registered as the reference image Ri without taking time and effort.

  In the above-described embodiment, the reference image Ri is registered in the digital camera 1, but the reference image Ri may be registered by connecting the digital camera 1 to a personal computer and operating the digital camera 1 from the personal computer. Good. Further, the reference image data Ri to which the identification information is added is recorded on the medium, the medium is set in the image recording unit 10, and the reference image data Ri recorded on the medium is stored in the reference image storage unit 6. A reference image may be registered.

  In the above embodiment, the reference image data Ri is registered with a person as a subject, but the subject is not limited to a person, and may be a character in a building or a theme park.

1 is a schematic block diagram showing the configuration of a digital camera according to an embodiment of an image pickup apparatus of the present invention. The flowchart which shows the process performed at the time of imaging | photography of the digital camera by this embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Digital camera 2 Imaging part 4 Image memory | storage part 6 Reference image memory | storage part 8 Judgment part 10 Image recording part 12 Display part 14 Input part

Claims (4)

Reference image storage means for storing a plurality of reference images each including a plurality of types of subjects in association with identification information for identifying subjects included in each of the plurality of reference images;
Imaging means for acquiring an image representing a subject by photographing;
A determination means for calculating a similarity between each of the acquired images acquired by the photographing and the plurality of reference images, and determining whether the similarity is equal to or higher than a predetermined threshold;
An image imaging device comprising: an information adding unit that adds the identification information associated with the reference image determined to be similar to the acquired image to the acquired image when the determination is affirmative apparatus.   The image capturing apparatus according to claim 1, wherein the subject is a person, and the identification information is information for identifying the person. Obtain an image representing the subject by shooting,
Reference image storage means for storing an acquired image acquired by photographing and a plurality of reference images each including a plurality of types of subjects in association with identification information for identifying subjects included in each of the plurality of reference images. Calculating similarity with each of the stored reference images;
Determine whether the similarity is greater than or equal to a predetermined threshold,
When the determination is affirmative, the identification information associated with the reference image determined to be similar to the acquired image is added to the acquired image. A procedure for acquiring an image representing a subject by shooting;
Reference image storage means for storing an acquired image acquired by photographing and a plurality of reference images each including a plurality of types of subjects in association with identification information for identifying subjects included in each of the plurality of reference images. Calculating a similarity with each of the stored reference images;
A procedure for determining whether the similarity is equal to or greater than a predetermined threshold;
And a step of assigning the identification information associated with the reference image determined to be similar to the acquired image to the acquired image when the determination is affirmed. A program to make it run.

Images