JP2008103850A - Camera, image retrieval system, and image retrieving method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a camera, an image retrieval system, and an image retrieving method that can easily retrieve what a photographed subject is by taking information on the size of the subject etc., into consideration whatever shape the photographed subject has. <P>SOLUTION: An MPU 41 decides a theme of an image photographed by the camera and the size of a main subject in the image. Information on the decision is recorded in a database 56a for image retrieval while related to the image. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a system, a service and an image search method for quickly searching for a photo by applying image search, and a camera that can take such a photo.

  Various attempts have been made to search for a necessary image from a vast number of images, many of which have been searched based on the date of a photo, keywords, and the like. There is also known a technique in which a digital watermark is added to an image itself and retrieval is performed with reference to the digital watermark (for example, see Patent Document 1 below).

Furthermore, there is also known one that attaches size information to an image by moving a cursor on the screen (for example, see Patent Document 2 below).
JP 2002-73618 A Japanese Patent Laid-Open No. 2004-40689

  However, as described in Patent Document 1, an image in which search data is included in an image as a digital watermark can only be used in a predetermined system based on the assumption, and the format of the digital watermark There is also a problem of how to enter data in

  In addition, as in Patent Document 2 described above, if the cursor is displayed on the screen every time shooting is performed, the rapid photographing property is insufficient. In addition, the application of the present invention in which not only measuring the size but also effectively utilizing such information to further expand the world of image appreciation is not possible.

  By the way, in recent years, with the advancement of IT technology, it has become possible to easily handle not only characters such as essays but also relatively fine images as important expression means in technologies based on the Internet, such as websites and blogs. , It is getting recognized by the market. And in these places, there are an increasing number of enthusiasts exhibiting boast photo works.

  On the other hand, many sites and providers have devised how to easily reach information accessible via the Internet. For example, not only search by characters but also a technique for searching for an image by determining the characteristics of the image itself has been studied and put into practical use. In other words, how to connect a user who wants to see an image and a user who wants to see an image is being studied by the industry.

  Under such circumstances, there has been a demand from the camera side for an effort to make an image easy to search and a technique for improving the system by considering both ideas.

  Therefore, the present invention has been made in view of the above circumstances, and the purpose of the present invention is to determine what the subject is in consideration of the size of the subject, regardless of the shape of the subject. To provide a camera, an image search system, and an image search method capable of facilitating a search for a certain item.

  That is, the invention described in claim 1 records the theme of the photographed image, determination means for determining the size of the main subject in the image, and information determined by the determination means in association with the image. And an associating means.

  According to a second aspect of the present invention, in the first aspect of the invention, the determination unit determines the width of the screen from the subject distance at the time of shooting and the shooting angle of view, and information about the width and the information at the time of shooting. The size is determined based on the focal length of the photographic lens.

  The invention according to claim 3 is the invention according to claim 2, wherein the determining means determines the size in consideration of whether or not the subject distance is in a macro area. To do.

  According to the fourth aspect of the present invention, the second photo group limited from the first photo group based on the feature information of the image and the size information of the main subject recorded in association with the image. It is characterized by selecting.

  According to a fifth aspect of the present invention, there is provided identification means for identifying a captured image based on a predetermined condition, and association means for associating information on the image identified by the identification means with the image. It is characterized by doing.

  The invention described in claim 6 is the invention described in claim 5, further comprising recording means for recording information of the image associated by the associating means.

  According to a seventh aspect of the invention, in the invention of the sixth aspect, the identifying means identifies the theme of the photographed image by comparing the size of the main subject in the image. It is characterized by.

  The invention according to claim 8 further comprises calculation means for calculating a screen width from the subject distance and the shooting angle of view in the invention according to claim 7, wherein the identification means The size is identified based on the information on the width of the screen calculated by the calculation means and the focal length of the photographing lens at the time of photographing.

  The invention according to claim 9 is the invention according to claim 7, wherein the theme of the image identified by the identifying means is identified by whether or not a face exists in a main subject in the image. It is characterized by that.

  According to a tenth aspect of the present invention, in the invention according to the ninth aspect, the theme of the image identified by the identifying means is the size of the face when a face exists in the main subject. It is characterized by being identified accordingly.

  The invention according to claim 11 is the invention according to claim 9, wherein the theme of the image identified by the identifying means is subject distance and subject color when the main subject does not have a face. It is distinguished according to at least any one of these.

  According to a twelfth aspect of the present invention, in the invention according to the sixth aspect, the identifying means compares the size of the main subject in the image according to whether or not the subject distance is in the macro area. It is characterized by identifying.

  According to a thirteenth aspect of the present invention, in an image retrieval system that retrieves an image from a plurality of images and obtains a desired image, it is based on the size information of the image identified based on a predetermined condition. Then, an image group corresponding to the image feature information is selected.

  The invention described in claim 14 is an image search system for obtaining a desired image by retrieving an image from a plurality of images, and an input means for inputting image feature information to obtain the desired image; Identifying means for identifying the plurality of images based on a predetermined condition; associating means for associating information on the image identified by the identifying means with the image; and information on images associated with the associating means; And extracting means for extracting the corresponding image group by comparing with the feature information of the image input by the input means.

  According to a fifteenth aspect of the present invention, in the invention according to the fourteenth aspect, the identification means compares the theme of the image with the size of the main subject in the image to identify the image. To do.

  According to a sixteenth aspect of the invention, in the invention of the fifteenth aspect, the theme of the image identified by the identifying means is identified by whether or not a face exists in a main subject in the image. It is characterized by that.

  The invention according to claim 17 is the invention according to claim 16, wherein the theme of the image identified by the identifying means is the size of the face when the main subject has a face. It is characterized by being identified accordingly.

  The invention according to claim 18 is the invention according to claim 16, wherein the theme of the image identified by the identifying means is subject distance and subject color when the main subject does not have a face. It is distinguished according to at least any one of these.

  According to a nineteenth aspect of the present invention, in the image search method for obtaining a desired image by searching for an image from among a plurality of images, a first image for identifying the plurality of images based on a predetermined condition is provided. A second step of associating the image information identified in the first step with the image, a third step of inputting image feature information to obtain the desired image, and the second step A fourth step of comparing the image information associated in the step and the image feature information input in the third step to extract a corresponding image group, and To do.

  According to the present invention, regardless of the shape of the photographed subject, it is possible to easily search for the subject by taking into account the size information of the subject and the like. A search system and an image search method are provided.

  In addition, image search is facilitated by combining such ideas without relying on one means of image determination. Further, in the system to which the present invention is applied, it is possible to construct an image retrieval system that can perform retrieval more efficiently, at high speed and accurately by applying such a concept.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

(First embodiment)
It is difficult to search for a desired image from a blog or a homepage that can be connected on the Internet. When performing an image search, in the database configuration for classification and search, such as facet classification, for example, the appearance, size, and color are important search keywords in order to identify things. At this time, it is possible to read the appearance and color from the image, but regarding the size, the human appearance and the description of the camera lens are so different from each other, so that some devising is required.

  Therefore, in the present invention, the size of the subject is determined and the information is also taken into consideration so that the search does not fail.

  FIG. 2A is a diagram conceptually showing a method of analyzing an image, determining its feature, and searching for an image. That is, when an image to be determined is input to the image determination, the shape of the image is determined by contour detection. At the same time, the color of the object is determined, and related keywords are searched from a huge database.

  However, in this case, a similar shape or color is searched for a picture drawn or a similar figure drawn on a poster behind the main subject instead of the main subject. For this reason, in the present invention, a system as shown in FIG. 2B is configured.

  In other words, auxiliary information (here, typical shape data, size data) etc. is added to the image to assist the search, and on the image determination side, not only the shape and color so far, but also the size Judgment is made in consideration of information and the like.

  Assuming such a configuration, as shown in FIG. 3A, the photograph 12 taken by the user 10 with the camera 11 to which the present invention is applied has attached data 12a. It is assumed that the photograph 12 is posted on a blog or the like from the input / output unit 13 via the network 15 and the provider 16. When a person 18 who wants to see the photograph 12 has an optimum keyword and additional information and searches on the net, the blog posted image 12 ′ can be surely viewed on the image search site 17. .

  On the other hand, if information used at the time of image search is input to the camera 11 from the site 17 where image search is performed, a design or the like designed for the typical image is synthesized and used as auxiliary information at the time of search. can do.

  In addition, know-how of image determination is changing every day, and not all image search sites provide such information in real time. Therefore, the feature is extracted from another provider 20 based on the plurality of images that are searched and displayed on the current search screen 19. A design pattern 21 ′ obtained by simplifying the typical image information 21 obtained in this way may be attached to an image photographed by the camera 11 or may be displayed in a composite manner.

  Since an image attached with such a design symbol is attached with an auxiliary image assuming a good pattern of a search engine, the system shown in FIG. 3A has a high hit rate at the time of search. .

  As shown in FIG. 4A, when the camera 11 is held at the time of taking a picture (observation on the monitor (display unit) 24 while reproducing the output of the image sensor), the release button ( The target information that the user 10 wants to shoot can be obtained from the contrast information obtained from the subject, the color, the background, the focus information, etc.

  FIG. 1 is a block diagram showing the basic configuration of a digital camera according to the first embodiment of the present invention. The digital camera (hereinafter abbreviated as “camera”) 11 includes a microcontroller (hereinafter referred to as “MPU”) 41 having functions of a determination unit, an identification unit, and a calculation unit, an autofocus (AF) control unit 42, and an aperture control. Unit 43, auxiliary school light emitting unit 44, photographing lens 46, aperture 47, image sensor 48, analog front end (AFE) unit 49, image processing circuit 50, contrast determination unit 51, color determination unit 52, face detection unit 53, compression unit 55, recording control unit 56, recording medium 57, input / output unit 58, display control unit 60, display unit 24, and a plurality of switches 23a, 23b, 23c. It is comprised with the operation part etc. which consist of.

  The MPU 41 that also functions as an associating unit and an extracting unit detects various operations by the user based on the states of the switches 23a, 23b, and 23c. The MPU 41 sequentially controls each block described above at the time of shooting according to the detection results of the switches 23a, 23b, and 23c and a predetermined program. The MPU 41 is responsible for overall control of the camera 11 such as shooting and playback according to a program.

  The switches 23a, 23b, and 23c notify the MPU 41 of a photographer's instruction. The switches 23a, 23b, and 23c are illustrated as representative examples of the operation unit. However, the switch 23a is a release switch, the switch 23b is a selection switch, and the switch 23c is, for example, switching between a shooting / playback mode and a shooting mode / display mode It is a switch.

  Further, the reading of the model image, which is a feature of the present invention, is also performed by detecting changes in these switches. For example, in the example shown in FIG. 2A, the camera 11 is framed, and theme determination and selection are performed from the composition at the moment when the selection switch 23b provided on the back of the camera 11 is pressed.

  An image from the subject 35 is received through an imaging lens 46 and a diaphragm 47 by an imaging device 48 as an imaging means constituted by a CCD, a CMOS sensor, or the like including a large number of light receiving surfaces (pixels). Since this is converted into an electrical signal in the image sensor 48, the result is converted into a digital signal by the AFE unit 49 including the A / D conversion unit and further input to the image processing circuit 50.

  The AFE unit 49 is also provided with a function for reading a plurality of pixels constituting the image sensor in a batch. For example, when the signal level of each pixel is small, such as 4 pixels (2 × 2) and 9 pixels (3 × 3), several pixel signals can be added to improve the S / N. Such an operation can increase sensitivity. As described above, when the sensitivity is increased, it is possible to devise photography such as increasing the shutter speed. When the shutter speed is high, it is possible to shoot as if the moving subject is stopped.

  In the camera 11 in which the user can switch the shutter speed and the appropriate exposure is controlled by controlling the aperture and sensitivity, if the shutter speed is generally slowed down, for example, when shooting a river flow, The extended description is possible. If the shutter speed is increased, even if the same river flow is photographed, the moment can be captured, and the moment when the water splashes and the state of ripples can be reproduced in detail. Even in sports photography, if the shutter speed is increased, the movement of the subject can be stopped and the expression can be clearly captured.

  As described above, how to set the shutter speed is also valuable know-how at the time of shooting, and is a topic to be shared among friends through the network as the present invention aims.

  The photographing lens 46 is for forming an image of the incident subject 35 on the image sensor 48. The photographing lens 46 includes an autofocus lens (focusing lens), and the focus position is switched by the MPU 41 via the AF control unit 42 according to the autofocus control.

  A mechanism (aperture 47) or the like that provides an effect of a shutter or an aperture is included in or near the photographic lens 46. This opens up to a predetermined aperture at the time of photographing, and closes when the exposure is completed to end the exposure. By changing the aperture 47, the depth of field of the photographic lens 46 changes, so it is possible to adjust the blurring of the background, etc. Switching can be done. In this way, the aperture value can also be learned through the network by looking at photos of friends.

  The AF control unit 42 controls the focus position of the taking lens 46 and monitors the focus position. The focus position control is executed when the image processing circuit 50 detects the contrast of the image data output from the image sensor 48 and outputs it to the MPU 41, and the MPU 41 outputs a control signal to the AF control unit 42 accordingly. . The MPU 41 outputs a control signal to the AF control unit 42 so that the contrast signal of the image data is maximized.

  The diaphragm control unit 43 is provided to control the diaphragm 47. As described above, the aperture control unit 47 performs exposure control to keep the amount of incident light on the image sensor 47 at a predetermined amount by changing the aperture 47.

  As described above, the image sensor 48 is composed of a CMOS or a CCD, and converts an image of a formed subject into an image signal. The AFE unit 49 converts the analog electrical signal output from the image sensor 48 into digital image data and outputs the digital image data. The AFE unit 49 also has a function of selecting signals output from the image sensor 48, and can extract a limited range of image data from all light receiving surfaces. In general, an image sensor can read out a pixel signal that is thinned out at a high speed, so that an image obtained for composition confirmation can be processed at a high speed and the display unit 24 can be connected via the display control unit 60. If displayed on the display screen, the user can perform framing as shown in FIG.

  For such a purpose, the image processing circuit 50 has a resizing unit (not shown), and the image sensor 48 displays the display screen of the display unit 24 made of liquid crystal or organic EL controlled by the display control unit 60. The signal is processed into a size that can be displayed there and input in real time. With this function, an image incident on the image sensor 48 can be confirmed prior to shooting, and the timing and photo opportunity at the time of shooting can be determined while looking at this instead of the viewfinder.

  As described later, since the image signal at this time can be used for focusing and the like, it has been described above that the focus position (distance) and the like can be used effectively when the theme is specified. By using the image at this time and performing image determination to be described later, it is possible to determine the scene that the user is going to shoot.

  Further, when the photographing lens 46 is a zoom lens, the distance can be determined in consideration of the zoom position and the like. Further, by adding and using several adjacent pixel signals of the same color of the image sensor 48, it is possible to improve the S / N and change the sensitivity. Such output addition processing can also be performed by the AFE unit 49. The image processing circuit 50 corrects the color, gradation, and sharpness of the input signal, and performs compression recording in a subsequent circuit during photographing.

  The image processing circuit 50 further includes a sensitizing unit 50a for amplifying the obtained image signal to a predetermined level and setting it to the correct shading and the correct gray level. The sensitizing unit 50a performs digital calculation so that the digitized signal level becomes a predetermined level.

  In addition, the image processing circuit 50 has an image determination function for determining the characteristics and the like of the image input from the image sensor 48 using the signal of the image processing circuit 50. For example, a contrast determination unit 51 that determines the contrast of an image signal level, a color determination unit 52 that detects a color component included in image information, and a face detection unit that detects the position of a main subject by examining image characteristics 53 etc. are provided in this system. This is for determining what kind of photograph the user is going to take when taking a picture. From this result, it is possible to determine the theme of the photo, or to determine the difference in background color even if the theme is the same. When a theme is determined, a typical image representing it is combined with the captured image as described above.

  Of these, the contrast determination unit 51 is used for autofocus control in conjunction with the focusing means of the photographic lens 46. As described above, the distance of the subject or the subject is determined by the lens position information at the time of focusing. The background distance or the like can be determined. Furthermore, it is possible to determine where the high contrast is.

  Further, the time when the photograph is taken is measured by the clock unit 41a in the MPU 41 and is associated with the image by the MPU 41. This shooting time information is very important, and it is possible to know the shooting theme from the shooting time and frequency (for example, a night view is a night view).

  In this way, signals and other information output from the image processing circuit 50 at the time of shooting are appropriately compressed by the compression unit 55 and recorded on the recording medium 57 at the subsequent stage.

  The signal output from the image processing unit 50 is compressed by the compression unit 55. In the compression unit 55, a still image compression unit 55a such as a JPEG core unit (shown as compression S in FIG. 1), MPEG4, H.264, or the like. It has a moving picture compression section 55b (shown as compression M in FIG. 1) composed of a compression core section such as H.264.

  The image signal and other accompanying data (theme information and size information) compressed by the compression unit 55 are recorded on a recording medium 57 as a recording unit via the recording control unit 56. The recording control unit 56 includes an image search database (theme determination database) 56a which is a recording unit. This content may be recorded when the camera is manufactured, but as described above, it may be rewritable in accordance with the search engine via the Internet.

  On the other hand, the captured image data recorded on the recording medium 57 can be uploaded to a service such as a homepage or a blog by the input / output unit 58 via the network 15, a personal computer 65 having an input means, and the like. In addition, when transmitting from the input / output unit 58, information indicating what theme the transmission photograph has can be transmitted. Of course, transmission is not limited to the Internet, and for example, a configuration in which transmission is performed via a memory card or a personal computer may be applied to the present invention.

  As described above, the information attached to the photographed image is theme information using time at the time of photographing, exposure control parameters and distance information, color information, face detection described later, and the like. As described above, this embodiment attaches not only the information of the image itself to be photographed and transmitted but also the information of the image related thereto, in particular, it is attached in the form of an image in addition to the case of a keyword. It is the feature.

  As will be described later, the image search system can make a correct determination based on such accompanying information. Since it has such a configuration, other net users 66 can know what kind of image the person is taking a picture of, and can deepen exchange.

  In addition, a plurality of images can be displayed on the display screen of the display unit 24 by using the multi-image display unit 60 a in the display control unit 60. This controls, for example, that an odd-numbered input image is displayed on the left side of the screen and an even-numbered input image signal is displayed on the right side of the screen. With this multi-screen display function, it is also possible to compare and view captured images.

  In addition to the main image 25, a candidate theme image (typical image) 26 can be displayed on the display unit 24 as shown in FIG.

  The user's operation is detected by the state of the switches 23a to 23c, and the MPU 41 determines that and sequentially controls each block at the time of shooting. In accordance with the operation of these switches 23a to 23c, shooting / playback mode switching, shooting mode switching, and the like can be performed. Also, as described above, reading of the model image, which is a feature of the present invention, is also performed by detecting changes in these switches. For example, in the example shown in FIG. 4A, the camera 11 is framed, and the theme is determined and selected from the composition at the moment when the switch 23b provided on the back of the camera is pressed.

  As an example of this theme selection, as shown in FIG. 5A, the keyword list table 72 is displayed on the display unit 24 by the operation of the switch 23 b by the user's hand 70. Among them, a corresponding keyword is selected by operating a selection switch 23b such as a cross key (not shown). In this state, it is conceivable that an operation associated with the image 71 is performed by operating the switch 23a.

  Information having a size characteristic of the present embodiment may be included in a keyword so that the information can be selected. In addition, the auxiliary light emitting unit 44 irradiates the subject 35 with light according to the situation to prevent insufficient brightness and uneven brightness.

  Next, the control operation of the camera having such a configuration will be described with reference to the flowchart of FIG. This control operation is performed by the MPU 41 based on a predetermined program.

  When this sequence is started, it is first determined in step S1 whether or not it is a photographing mode. The determination as to whether or not the camera is in the shooting mode is made by pressing the release button 23a for shooting. As a result, if it is the photographing mode, the process proceeds to step S2, and if not, the process proceeds to step S8. In step S8, it is determined whether or not the playback mode is set. If it is the reproduction mode, the process proceeds to step S18, and if not, the process proceeds to step S9.

  In step S9, it is determined whether or not a power switch (not shown) is turned off. As a result, if the power is turned off, a processing operation for turning off the power is performed. On the other hand, if the power is on, the process proceeds to step S10.

  For example, although shooting is started by pressing the release button 23a for shooting, an image to be shot is displayed on the screen of the display unit 24 before that. Then, the user confirms the subject to be photographed and the situation while looking at the display screen. At this time, the imaging & display process is executed in step S10 through step S1 to step S8 and further step S9. Here, the focus position of the photographic lens 46 is moved as necessary, the contrast determination unit 51 determines the contrast change of the image, and the autofocus operation is appropriately executed. That is, the distance from the focused lens extension position to the subject 35 is obtained by the AF control unit 43, the MPU 41, and the like.

  Next, in step S11, the contour of the main subject is detected from the image obtained in this state. This is done by judging whether or not the change points of the image raised by the edge enhancement technique of the obtained image are continuous, or by irradiating auxiliary light from the auxiliary light emitting unit 44 and detecting the reflected light distribution Or just make a decision.

  For example, as shown in FIG. 7A, when the auxiliary light is emitted from the auxiliary light emitting unit 44 to the subject 35 prior to photographing, the reflected light distribution as shown in FIG. The relationship between the angle of view θ and the distance L between the subject can be determined. The boundary between the place where the reflected light returns and the place where it does not come can be determined as the contour portion.

  In step S12, the size of the subject 35 is determined. The size of the subject is determined based on the ratio of the subject that can be seen from the contour portion (the subject width shown in FIG. 7B) in the screen, the angle of view of the photographing lens 46, and the distance information described above. It is possible. If the distance L and the angle of view θ of the subject 35 are known, the photographing range when photographing the subject 35 can be known, and the size of the subject 35 can be known by the ratio of the photographing range.

  In this way, depending on whether the subject is large or small, there are several candidates for what the subject is based on the concept and image characteristics as shown in the flowchart of FIG. 11 described later. Therefore, when the subject candidates are determined in step S13, the candidates are listed on the display screen of the display unit 24 in subsequent step S14. Then, in step S15, the user views the list display and selects a theme from the candidates.

  Here, when a candidate is selected, according to the operation, theme information can be given to an image to be taken in step S16. Thereafter, when size information is given in step S17, the process proceeds to step S1.

  In step S1, when shooting is started by pressing the release button 23a for shooting or the like, exposure and AF control are performed again in step S2. Next, time determination is performed in step S3. In step S4, it is determined whether the theme has been set.

  Here, when the theme determination (details will be described later) has been made from the imaging result that has been performed previously, or as shown in FIG. 4A, the theme setting selection switch 23b is operated and the theme is selected. If it is photographed, the process proceeds to step S5 where photographing and storage are performed. In step S6, the theme information association such as the theme information and size information is compressed and recorded. Thereafter, the process proceeds to step S1.

  On the other hand, if the image is taken without setting the theme in step S4, the process proceeds to step S7, where normal shooting is performed. The captured image is compressed and then recorded on the recording medium 57. Thereafter, the process proceeds to step S1.

  In a situation where no shooting operation is performed, the process proceeds from step S1 to step S8 to determine whether or not the playback mode is set. If the playback mode is set by the user, the process proceeds to step S18. Playback display is performed. In this process, what is selected by the user from the photographing result is reproduced and displayed on a display screen such as a liquid crystal display. At this time, the theme may be set for the reproduced image. When this is done, it is determined and implemented in step S27.

This theme setting method is as follows. For example, as shown in FIG. 8A, when the display panel of the display unit 24 is configured with a touch panel, the user selects a desired image from the reproduced captured images 24a. Then, as shown in FIG. 8 (b), to list 24b themes to be associated, may be the photographed image 24a ₁ selected so as to associate in such a manner as to drag a finger. In such a system, since the relationship between the displayed image and the theme can be easily and intuitively recognized, anyone can easily operate.

  Further, the size of the subject may be input by applying such a method. In other words, a size list is output on the display unit 24, and for example, numerical frames such as “several centimeters”, “ten centimeters”, “tens of centimeters”, “1 m”,. A device may be devised in which this is dragged and linked to the corresponding image.

  Furthermore, when performing theme setting and image upload, which are features of the present embodiment, the upload described above is performed when the image is selected for transmission from the image playback mode. That is, when it is desired to announce the photographed result, the fact is specified. That is, an image is selected to be reproduced in step S19, and the image selected in step S20 is displayed on the display unit 24. On the other hand, if there is no selected image, the process proceeds to step S8.

  It is determined whether or not the image displayed in step S20 is transmitted in step S21. As a result, when transmitting, it transfers to step S22, and when not transmitting, it transfers to step S26.

  In step S22, the presence / absence of theme information is determined. Here, when there is no theme information, it transfers to step S23, and after related information is provided, it transfers to step S25. On the other hand, if there is theme information in step S22, the process proceeds to step S24, and related information is given. Thereafter, in step S25, it is transmitted outside the camera. This transmission may be a transmission via a personal computer via a USB cable, a LAN cable, a wireless LAN, or infrared communication. This transmission setting and the like are also performed using the selection switch 23b and the like. After the transmission in step S25, the process proceeds to step S8.

  If it is not transmitted in step S21, the process proceeds to step S26 to determine the presence / absence of theme setting. Here, when the theme setting is performed, the process proceeds to step S27, the theme information is associated, and then the process proceeds to step S8. On the other hand, if the theme is not set in step S26, step S27 is skipped and the process proceeds to step S8.

  Next, the principle of face detection by the face detection unit in the first embodiment of the present invention will be described with reference to FIG. 9 and FIG.

  FIG. 9 is a diagram showing an example of a face shadow comparison pattern used when face detection is performed. As described above, a plurality of face patterns 28a, 28b, and 28c having different sizes are scanned on an image as shown in FIG. 10A at high speed, and the degree of coincidence with the image is compared. It is possible to know whether the position is the highest comparison result and the degree of coincidence is high. The result can be determined as face size and position information as shown in FIG. A template corresponding to the shadow of a human face as shown in FIG. 9 is changed in size as shown as face patterns 28a, 28b, and 28c in FIGS. 9 (a), (b), and (c). As shown in FIG. 10A, it is determined that a face is present where the degree of coincidence is high. In this case, the face position is determined by the face pattern 28b.

  Thus, the position and size of the face are specified as a circle 31a as shown in FIG.

  Here, an outline of how to determine the theme of a photographed image using the above-described function of detecting a face will be described with reference to the flowchart of FIG.

  When this sequence is started, in step S31, it is determined whether or not a face is included in the image obtained from the image sensor 48. If there is a face, the process proceeds to step S32, and the size is compared with a predetermined size. By this comparison, the second photo group is selected from the first photo group. As a result of the comparison in step S32, if it is small, the process proceeds to step S33 to select a snapshot. On the other hand, if it is larger than the predetermined size, the process proceeds to step S34, where it is selected to be a portrait photograph.

  If it is determined in step S31 that the scene has no face, the process proceeds to step S35, and the distance is determined from the focusing result. If it is determined that the distance is a long distance, the process proceeds to step S41 to determine that it is a landscape photograph. If it is determined in step S35 that the distance is not a long distance, the process proceeds to step S36 to determine whether the distance is a close distance. Here, if it is determined that the distance is not close, that is, if the distance is close or far, the process proceeds to step S40 to be a pet photograph.

  If it is determined in step S36 that the distance is close, the process proceeds to step S37, where color determination is performed. In this color determination, the shooting screen is divided into a plurality of, for example, nine, a green subject (located in the G component region in the CIE chromaticity diagram) exists in the periphery, and a subject other than green ( That is, it is determined whether or not there is a CIE chromaticity diagram located outside the G component region. As a result, what meets these conditions is determined to be a photograph of a flower by moving to step S38. On the other hand, if it is determined in step S37 that the color condition described above is not met, the process proceeds to step S39, where the photograph is a small accessory such as an accessory.

  Of course, the color of a photograph such as a landscape may also be determined. For example, the sky or the sea may be determined depending on where there is a blue object, and dusk may be determined if the color is orange.

  As described above, when the outline is determined, a more detailed keyword may be selected by the user from a predetermined list. In addition, the user may set what kind of theme to classify without going through such a process. However, it is possible to perform a quick judgment and operation even when attaching more detailed keywords after performing such a narrowing down.

  In such a camera, in a situation where the camera is held as shown in FIG. 4A, the pre-shooting image 25 as shown in FIG. An image 26 of a theme that is a candidate can be inferred from the distance and size of the subject at that time, and can be displayed simultaneously. The user may select it from the candidates by a selection method using a cross key or a jog dial. The selection result may stick to the image like a keyword, but here, for example, as shown in FIG. 4C, a simple image is synthesized by synthesizing it as a typical image at the lower right of the image 25. Even if the similar image search technique is applied, this image can be easily selected during the search.

  However, if this attached composite image is large, it will interfere with the actual image, so it must be displayed in a reduced size. In addition, if it is displayed in a reduced size, it will be inaccurate if the size is calculated from the angle of view and distance. For images with this attached composite image, the focal length information, angle of view information, and distance information are invalidated. Is preferred. In addition, if the image is resized or trimmed, the size of the subject cannot be analogized. Therefore, if size information can be obtained at the time of shooting or at a timing preceding it, it is preferable that the information can be recorded.

  Next, the function of the provider that performs image search as assumed in the present embodiment will be described with reference to the flowchart of FIG.

  First, in step S51, it is determined whether a photographic image has been sent from the user. If an image has been sent, the process proceeds to step S52, and information attached to the image is determined. Next, in step S53, posting is performed on an appropriate homepage or blog screen. Then, in step S54, it is determined whether or not the image has theme information. As a result, if it is an image with theme information, the process proceeds to step S55, where appropriate display is performed, a signal is added, or the image display method is changed when the image is displayed. Or In step S56, the size information of the subject is recorded in the same manner as the keyword in association with the image so that it can be easily searched later. Thereafter, the process proceeds to step S51.

  If it is determined in step S51 that the image has not been received from the user, the process proceeds to step S57 to determine whether the search is a theme. Here, when a net user searches for a predetermined theme, the process proceeds to step S58, and a blog or an image having the keyword of the theme is determined. If there is a corresponding item, the process proceeds to step S59, and the corresponding blog or image is displayed. At this time, when subject size data or the like is attached to the image, a similar theme that can be assumed is searched from the information.

  In other words, for example, if the theme of the blog search is a dog or cat pet, we will focus on those with a size of several tens of centimeters to 1.5 meters, and will not search for information with a smaller size, Also, the larger ones are useless so do not search. This improves the accuracy of the image search. However, since the information such as the color, shadow, and shape of the image can be used effectively, it is possible to efficiently search by making use of the features of the image search and supplementing at least the portion that cannot be expressed by the feature.

  In step S58, if there is no search target with information on the corresponding theme, step S59 is skipped. In step S60, referring to the content of the image itself and the size of the subject, an image including a typical image to be synthesized is searched. Next, in step S61, a blog-like homepage (HP) having the same is displayed. In addition, information on typical themes handled by the provider or site can be transmitted to the camera.

  If this information is used, the representative theme and keywords can be displayed in the theme list part as shown in the list 72 in FIG. 5A, and can be easily associated with the blog.

  If it is determined in step S57 that it is not a theme search, the process proceeds to step S62, and the presence / absence of theme information is determined. Here, when there is no theme information, it transfers to said step S51, and when there exists theme information, it transfers to step S63. In step S63, it is determined whether or not a list is displayed.

  When the user wants to input this representative theme to the camera and sets that on the personal computer or the like, the process proceeds to step S64, the theme list is transmitted, and information is output in step S66. A display list indicating the size of the subject may be included in this theme. Since the concept of this size classification also changes depending on the search method, it cannot be generally determined only from the camera side, and thus a device such as a recommended input method from such a provider is effective.

  In addition, a user who is pursuing only a predetermined theme may acquire typical image data corresponding to the theme. In this case, if this is set while viewing this search site on a personal computer, the process proceeds from step S63 to step S65, and the typical image can be input to the camera. For example, in the case of a mammal pet such as a dog or a cat, this is a clear image that is easy to search, such as a quadruped or an ear in the house of the head.

  As described above, the photograph taken with a camera that can synthesize a typical image according to the theme according to the concept as in the present embodiment is based on the determination method and the theme used by the search service on the Internet. Can be easily classified, and there is no omission in searching.

  In addition, in a site that places importance on blog posting instead of a site that places importance on search as shown in the flowchart of FIG. 12, as shown in the flowchart of FIG. If the feature can be determined, it is possible to facilitate the image search at the assumed search site.

  In the flowchart of FIG. 13, the processing operations in steps S71 to S75 are the same as those in steps S51 to S55 in the flowchart of FIG. 12 described above, so that description will be given here with reference to the corresponding step numbers. Is omitted.

  If it is determined in step S71 that the image has not been received from the user, the process proceeds to step S76, and it is determined whether or not information on the assumed search site has been acquired. Then, when the information on the assumed search site is acquired, the process proceeds to step S77, and the theme to be picked up for the search service selected by the user is selected. Then, in the subsequent step S78, the characteristics (shape, color, etc.) of the subject in the image posted there are classified and extracted.

  Furthermore, when an image satisfying the extracted features is created in step S79, the theme and image information are associated in a subsequent step S80. Next, in step S81, these pieces of information are transmitted to the camera together. In this way, it is possible to synthesize an image as shown in FIG. 4C from an image example displayed as a list on the site 19 shown in FIG. It is possible to obtain a target image.

  As described above, according to the present embodiment, it is possible to provide a system that can quickly access accurate information using the direct expressive power of an image and the size information of a subject. For example, the LP and CD jackets are the same if only the image data is used, but the difference between them can be appropriately determined by including size information.

  As shown in FIG. 14, the image in the memory (recording medium) 57 of the camera 11 can be obtained by simply connecting the camera 11 to the personal computer 75 connected to the network 15 via the connector (input / output unit) 58. A system in which the image 71 set to be transmitted to the blog provider 16 is transmitted, or the theme list 72 handled in the blog is input from the provider 16 to the memory 57 of the camera 11 and the user is associated therewith. Good. In such a system, if the provider 16 performs control according to the flowchart as shown in FIG. 15, for example, an image displayed on the display unit 24 of the camera is displayed on the network connection means such as the personal computer 75 via the connector 58. Just connect to, you can get a blog posting and a theme list.

  With such a device, it is possible to provide an easy-to-operate camera that can update the latest themes handled by blogs and sites, and can upload images only by connecting to the Internet.

  First, in step S91, it is determined whether or not the camera 11 is connected to the personal computer 75. Here, if both are connected, in the subsequent step S92, it is determined whether or not there is an upload candidate image at that time. As a result, if there is no candidate image, the process proceeds to step S97 described later, and if there is a candidate image, the process proceeds to step S93. In step S93, the attached information attached to the image is determined, and in step S94, the blog is posted according to this information.

  Next, in step S95, it is determined whether the image is a theme image. If it is a theme image, the process proceeds to step S96, and theme information (for example, a typical image mark or the like) is attached to the image. On the other hand, if it is not the theme image in step S95, step S96 is skipped. In this state, the theme information handled by the blog or site is transmitted to the camera 11 in step S97. In steps S98 and S99, a typical image mark (typical image information) representing the theme is transmitted.

  As described above, according to the present embodiment, it is possible to provide a camera having high affinity with a blog, a homepage, or the like.

  In addition, the size determination method for estimating what the subject is has been described with reference to FIG. 7, but the method is not limited to this, and the configuration shown in FIGS. You may make it determine the magnitude | size of.

  FIG. 16A is a diagram illustrating a state in which the camera is held with respect to the subject 35 and the focus adjustment unit of the photographing lens 46 is moved to monitor the output change of the image sensor. In this case, if there are a plurality of contrast change detection units (not shown) in the imaging unit as shown by white circles in FIG. 16B, the contrast change at each point is shown in FIG. 16C. Obtained as shown. In this change, since the peak position changes depending on the object distance, if there is a peak at the same position in the black circle part of FIG. 16B, it can be seen that there is an object of the same distance in this part and the object is in this range. .

  In this way, it is considered that the main subject exists at a point where the contrast change state is similar to that at the center of the screen, and the size of the subject can be determined from the size occupied in the screen.

  In addition, as shown in FIG. 17, in a system including another sub lens 46a and the image sensor 48a in addition to the main photographic lens 46 and the image sensor 48, an image obtained by an observation system with parallax is obtained. The subject distance can be determined by the principle of triangulation. Therefore, the size of the subject 35 can be determined from the range of the part having the same distance.

(Second Embodiment)
Next, a method for measuring the size of a subject by a camera and attaching information to an image will be described as a second embodiment with reference to FIGS.

  In the second embodiment, the basic configuration of the camera is the same as that of the first embodiment shown in FIGS. 1 to 17, and the basic operation is also the same. Therefore, for these configurations and operations, the same reference numerals are assigned to the same parts, and illustration and description thereof will be omitted, and only the configurations and operations of different parts will be described.

  FIGS. 18A and 18B are diagrams showing an example in which the subject 35 is photographed at a short distance Lm. For example, as shown in FIG. 18A, in a situation where a wide range of the subject 35 is desired to be displayed on the full screen as much as possible, such as a fish dish on the plate 36, the user shortens the focal length of the zoom lens 46a. (Wide-angle side: Wide) tends to shoot. This is because, in a situation where the user wants to shoot the entire image at the same time and also wants to shoot the details well, the user performs such shooting by devising the distance. In this case, rather, the angle of view at the time of shooting is fully expanded, and shooting is performed by changing the distance.

  Normally, in terms of a film camera, a lens starting from a focal length equivalent to 35 mm can capture an image sensor having a width of 36 cm at a distance of 35 cm and a width of 72 cm at a distance of 70 cm. The change in the distances of 36 cm and 70 cm can be adjusted by the photographer while changing the composition. Rather, it can be controlled more easily than operating the zoom lens.

  In contrast to this, when photographing a fish or the like swimming, it is difficult to enter a distance that hinders the activity. In this way, when the opponent is a living thing, the approach of the photographer with the camera will have a positive effect, so the distance at the time of shooting is determined first, so as not to interfere with the activities of the living thing. It is more common to determine the composition and timing from a relatively far distance. Therefore, there are many cases where the image is taken on the telephoto side (Tele side) of the zoom lens. Moreover, in the case of a small animal having a length of several centimeters, this tendency is remarkable, and this tendency becomes remarkable because it is desired to photograph as large as possible.

  For example, in the above-described example, in terms of a film camera, a lens starting from a focal length of 35 mm corresponds to a width of the imaging element of 36 mm. Therefore, if the lens has a width of 3.5 cm, the distance is 3.6 cm and the width is 7.2 cm. If you want to take a full picture of it, you must shoot at a distance of 7.2 cm. On the other hand, with 3x and 5x zoom, etc., a 20cm to 35cm degree of freedom is born, so you can enjoy much more realistic and reasonable shooting compared to unreasonable shooting close to a distance of several centimeters. be able to.

  These small creatures remind you of fresh emotions just by magnifying and shooting. If you can record a sight that is different from what you usually see, you can shoot a magnified image, and it can be fully produced depending on the angle and timing of the shooting (even if there is no background, etc.). .

  Furthermore, if the subject is not a small subject as described above but is about the size of a human being, a dog, or a cat, as shown in FIG. It is possible to shoot at an appropriate size in the screen at the distance Ln. These are orders of magnitude larger than the fish of the order of several centimeters mentioned above, and even if they are sights that you normally see with the naked eye. As a way to improve the appearance, the background to be combined is also an important factor. However, as long as it is the main subject, it is assumed that it is contained within a certain size within the screen.

  With this concept, as shown in FIG. 18A, when the width X of the area to be photographed and the distance L of the subject are illustrated, from the focal length f of the photographing lens and the width W of the image sensor, FIG. A graph as shown in FIG. In the figure, when the focal length f is changed, it is switched to W (wide angle) (short focal length) or T (telephoto) (long focal length).

  In the case of a short distance called macro photography, a photograph as shown in FIG. 20 (a) or a photograph as shown in FIG. 20 (b) is obtained. It has already been described with reference to a) and (b). That is, a subject with a size of several centimeters is photographed on the telephoto side, and a subject with a size of several tens of centimeters is photographed on the wide angle side.

  Also, what is photographed at a distance called infinity is a building or the like as shown in FIG. 20 (e), and tends to be photographed on the wide-angle side, as shown in FIG. 20 (f). As you can see, a mountain-like object is often photographed on the telephoto side in order to draw it closer to you. From such a feature, it is possible to distinguish and determine a thing on the order of several tens of meters and a thing on the order of several kilos depending on the focal length.

  On the other hand, what is not at infinity and is located at a distance between them is macro photography, as shown in FIGS. 20 (c) and 20 (d), what is photographed on the telephoto side is taken by the whole body of the subject. It is thought that. In addition, it is possible to determine that the subject photographed on the wide angle side is photographed together with the landscape.

  When such a concept is used, the approximate size of the subject is determined based on a concept that further considers the zoom position with respect to the shooting range X obtained from the focal length f and the distance L according to the flowchart shown in FIG. Is possible.

  That is, in step S101, the distance information L is obtained from the extended position of the focusing lens at the time of shooting. Next, in step S102, focal length information f is obtained from the zooming position. Further, in step S103, the width X of the portion that can be photographed on the screen is obtained from the relationship as shown in FIG. In step S104, based on such information, it is determined whether or not the distance of the subject is a macro area (about 1 m or less).

  Here, if the distance of the subject is the macro area, the process proceeds to step S105, and it is determined whether the focal length is the telephoto side (Tele side) or the wide angle side (Wide side). As a result, if it is the telephoto side, the process proceeds to step S106, and it is considered that the state is as shown in FIG. . This takes into account that the image is captured at a certain size in the screen. This is because it looks good when shooting a small subject with this size (a fraction of the size of the photo screen). If it is smaller than this, the eyes and mouth of the object are well depicted. Not.

  If it is determined in step S105 that the lens is not on the telephoto side, it is determined in step S107 whether the lens is on the wide angle side. As a result, when the zoom position at the time of shooting is on the wide angle side, the process proceeds to step S108. In this step S108, since the subject is to be stored on the full screen, the screen range X is considered as the size of the subject as it is.

  Furthermore, when it is determined in steps S105 and S107 that the focal length is other than that, the process proceeds to step S109. In this case, assuming that the telephoto side and the wide-angle side are combined, the width of half of the shooting range X is considered as the size of the subject.

  Further, the concept of the size of the subject in the vicinity of the infinity distance (scenery) described with reference to FIGS. 20 (e) and 20 (f) is based on this. If it is determined in step S104 that the shooting is not in the macro area, and it is determined in step S110 that the shooting is equivalent to infinity, step S105 and subsequent steps are executed. On the wide-angle side, the entire screen is the subject, while on the telephoto side, a part of the screen is the subject.

  Further, in the above steps S104 and S110, if the scene is neither a macro area shooting nor a long distance shooting, the processing operation after step S111 is executed. Again, depending on whether the zoom lens is set to the telephoto side or the wide-angle side, the subject size determination is switched, but here the degree of freedom in distance to the subject is large (in the macro area, it is too close) If the subject escapes and the scenery moves, it does not change even if I move a little), but when the telephoto side is set, the subject is large, and when the wide angle side is set, the background is taken into account, and the main subject is rather small (See FIGS. 20C and 20D).

  Accordingly, if it is determined in step S111 that the camera is on the telephoto side, if it is on the telephoto side, the process proceeds to step S112 and the photographing range is set as the subject size Xo as it is. On the other hand, if it is not the telephoto side in step S111, it is determined in step S113 whether or not it is the wide angle side. Here, on the wide-angle side, the process proceeds to step S114, and it is considered that the subject is photographed small (1/10 of X). Further, in steps S111 and S113, if the focal length is between the telephoto side and the wide angle side, the process proceeds to step S115 and is considered to have a size about 1/5 of X.

  Such a device makes it possible to determine at least whether the subject is several centimeters, several tens of centimeters, or over 1 meter. With such information, it is possible to perform a search with much higher reliability than simply determining the theme based on the characteristics of the image.

  As described above, the size information of the subject can be taken into account by such a method, and in some cases, by combining the methods described with reference to FIGS. Search is possible. Once an image is captured by a personal computer, etc., it can be freely trimmed and resized by image processing software, etc., so subject size information must be clarified at the shooting stage and associated with the image. It is preferable to keep it.

  According to the above-described embodiment, in consideration of such a concern, the image search technique is complemented, and it is useful for activating correct and effective communication on the network.

  Of course, the embodiments described above are not limited to image searches on the network, but images recorded on a huge capacity hard disk drive in a personal computer, flash memories that have been increasing in capacity in recent years, etc. It goes without saying that the images recorded in the above can also be easily retrieved by applying the present invention.

  Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the scope of the present invention.

  Further, the above-described embodiments include inventions at various stages, and various inventions can be extracted by appropriately combining a plurality of disclosed constituent elements. For example, even if some constituent requirements are deleted from all the constituent requirements shown in the embodiment, the problem described in the column of the problem to be solved by the invention can be solved, and the effect described in the column of the effect of the invention Can be extracted as an invention.

1 is a block diagram showing a basic configuration of a digital camera according to a first embodiment of the present invention. (A) is a diagram conceptually showing a method of analyzing an image, determining its characteristics, and searching for an image, and (b) is a diagram conceptually showing a method of searching for an image according to the present invention. (A) is a figure explaining the example which displays the image image | photographed with the camera in this embodiment on a site via a network, (b) demonstrates the example which obtains the image for a search from the image on a site. FIG. (A) is a diagram for explaining an example in which a user takes a picture, (b) is a diagram showing an example of an image on a display unit of the camera, and (c) is an example of a composite image displayed. It is a figure. (A) is a figure explaining the example of keyword assignment, (b) is a figure explaining the example of keyword search. 3 is a flowchart for explaining a control operation of the camera in the first embodiment of the present invention. It is explanatory drawing for determining the magnitude | size of a to-be-photographed object. It is a figure explaining the method of a theme setting. It is a figure explaining the principle of the face detection by the face detection part in the 1st Embodiment of this invention. It is a figure explaining the principle of the face detection by the face detection part in the 1st Embodiment of this invention. It is a flowchart for demonstrating the outline of a mode that the theme of the image | photographed image is determined using the function which detects a face. It is a flowchart explaining the operation | movement of the provider who performs an image search which the 1st Embodiment of this invention assumes. It is a flowchart explaining the operation | movement of another provider which performs the image search which the 1st Embodiment of this invention assumes. It is a figure explaining an example in which a camera is connected to a personal computer connected to a network via a connector, and an image is input to the camera from a theme list handled in a blog. It is a flowchart explaining the function of the provider who performs the image search at the time of connecting as shown in FIG. It is explanatory drawing for determining the magnitude | size of a to-be-photographed object. It is explanatory drawing for determining the magnitude | size of a to-be-photographed object. It is a figure which shows the 2nd Embodiment of this invention and showed the example which is going to image | photograph the object in a short distance. 6 is a graph showing the relationship between the width X of a captured area and the distance L of a subject. It is the figure which showed the example of the image image | photographed in the af point of the graph of FIG. 12 is a flowchart for explaining an operation of determining the size of a subject in the second embodiment of the present invention.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 11 ... Camera, 12 ... Photograph, 12a ... Attached data, 13, 58 ... Input / output unit, 15 ... Network, 16 ... Provider, 17 ... Image search site (site), 23a ... Release switch (release button), 23b ... Selection Switch, 24 ... Display unit, 28a, 28b, 28c ... Face pattern, 35 ... Subject, 41 ... Microcontroller (MPU), 41a ... Clock unit, 42 ... Autofocus (AF) control unit, 43 ... Aperture control unit, 44 DESCRIPTION OF SYMBOLS Auxiliary school light emission part 46 ... Shooting lens 47 ... Diaphragm 48 ... Imaging element 49 ... Analog front end (AFE) part 50 ... Image processing circuit 50a ... Sensitization part 51 ... Contrast judgment part 52 ... Color determination unit, 53 ... face detection unit, 55 ... compression unit, 56 ... recording control unit, 57 ... recording medium, 60 ... display control unit.

Claims (19)

A determination means for determining a theme of a captured image and a size of a main subject in the image;
Association means for recording the information determined by the determination means in association with the image;
A camera comprising:   The determination means determines the width of the screen from the subject distance and shooting angle of view at the time of shooting, and determines the size based on the width information and the focal length of the shooting lens at the time of shooting. The camera according to claim 1.   The camera according to claim 2, wherein the determination unit determines the size in consideration of whether or not the subject distance is in a macro area.   A second photo group selected from the first photo group is selected based on the image characteristic information and the size information of the main subject recorded in association with the image. Search system. Identification means for identifying a captured image based on a predetermined condition;
Associating means for associating the image information identified by the identifying means with the image;
A camera comprising:   6. The camera according to claim 5, further comprising recording means for recording information on the image associated by the associating means.   The camera according to claim 6, wherein the identification unit identifies the theme of the photographed image by comparing the size of the main subject in the image. A calculation means for calculating a screen width from the subject distance and the shooting angle of view at the time of shooting;
8. The camera according to claim 7, wherein the identifying unit identifies the size based on the information on the width of the screen calculated by the computing unit and the focal length of the photographing lens at the time of photographing. .   The camera according to claim 7, wherein the theme of the image identified by the identifying unit is identified by whether or not a face exists in a main subject in the image.   The camera according to claim 9, wherein the theme of the image identified by the identification unit is identified according to a size of the face when a face exists in the main subject.   10. The theme of the image identified by the identifying means is identified according to at least one of a subject distance and a subject color when no face exists in the main subject. Camera described in.   The camera according to claim 6, wherein the identification unit compares and identifies the size of a main subject in the image according to whether or not the subject distance is in a macro area. In an image retrieval system for retrieving a desired image by retrieving an image from a plurality of images,
An image search system, wherein an image group corresponding to the image feature information is selected based on image size information identified based on a predetermined condition. In an image retrieval system for retrieving a desired image by retrieving an image from a plurality of images,
Input means for inputting image feature information to obtain the desired image;
Identification means for identifying the plurality of images based on a predetermined condition;
Associating means for associating the image information identified by the identifying means with the image;
Extraction means for comparing the image information associated by the association means with the feature information of the image input by the input means and extracting a corresponding image group;
An image search system comprising:   15. The image search system according to claim 14, wherein the identification unit compares the theme of the image with the size of a main subject in the image for identification.   The image search system according to claim 15, wherein the theme of the image identified by the identification unit is identified based on whether or not a face exists in a main subject in the image.   The image search system according to claim 16, wherein the theme of the image identified by the identification means is identified according to the size of the face when a face exists in the main subject.   The theme of the image identified by the identifying means is identified according to at least one of a subject distance and a subject color when no face exists in the main subject. The image search system described in 1. In an image retrieval method for obtaining a desired image by retrieving an image from a plurality of images,
A first step of identifying a plurality of images based on predetermined conditions;
A second step of associating the image information identified in the first step with the image;
A third step of inputting image feature information to obtain the desired image;
A fourth step of comparing the image information associated in the second step with the image feature information input in the third step to extract a corresponding image group;
An image search method comprising:

Images