JP2008065792A - Image processing apparatus and method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To display images in various modes using a plurality of templates by an electronic photo album. <P>SOLUTION: Layout templates for previously setting layouts of images prepared. A layout template 231 is used for notifying a browsing person of a short imaging interval by arranging three images in an overlapped manner. A layout template 233 is used for notifying the browsing person of the long imaging interval by arranging the three images so as not to be overlapped and arranged in line. These layout templates are selected based on imaging intervals of the images to be arranged. The present invention can be applied to a device for providing the electronic photo album. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an image processing apparatus, method, and program, and more particularly, to an image processing apparatus, method, and program capable of creating an album in which the arrangement of images is determined according to the date and time when the image was taken.

  Digital cameras and the like have become widespread, and images taken with digital cameras can be easily browsed with a personal computer or the like.

  Before the popularization of digital cameras, cameras called so-called film cameras were generally used. In general, an image (photograph) taken by a film camera is browsed in a state of being printed on predetermined paper. Those photographs are sometimes managed in a form called an album or the like in a state of being affixed to the mount.

  It has been proposed to use an album that has been used when browsing images taken with such a film camera also when browsing images shot with a digital camera. An album that browses images taken by a digital camera may be referred to as an electronic album.

  The electronic album can be viewed on a display of a personal computer, a television receiver or the like. Also, an electronic album is generally composed of a plurality of pages so that it can be handled in the same manner as an album on paper media, and generally has a display form in which the next page is displayed when the page is turned. The user can enjoy the electronic album with the same feeling as the paper medium album.

The album of the paper medium is created by the user pasting the photo on the mount. In the case of an electronic album, it is possible to create an album by pasting an image by a user performing a predetermined operation on a screen that images a page displayed on the display. It is also proposed to create an album by placing images on the page instead. (For example, see Patent Documents 1 and 2)
Japanese Patent No. 3528214

JP 2001-228528 A

  When creating an electronic album without bothering the user, various functions are provided so that the user can enjoy the electronic album as well as creating it. In addition, it is required to display images using various expression methods.

  The present invention has been made in view of such a situation. When an image is displayed, the image can be displayed in various forms using a plurality of templates.

  An image processing apparatus according to an aspect of the present invention is a template for determining an image arrangement, and includes a management unit that manages a template prepared for each number of images to be arranged, and an image attached to the image to be grouped. Included in the group generated by the processing of the grouping unit, the grouping unit that divides the image into one or more groups based on the information acquired by the acquisition unit, and the grouping unit Selection means for selecting the template managed by the management means according to the number of images, and placement means for placing the image on the template selected by the selection means.

  The management unit manages a plurality of templates having different positions for arranging images for each number of the images.

  The grouping means groups so that the number of images included in one group is equal to or less than the maximum number that can be arranged by one template managed by the management means.

  The information may be information related to the date and time when the image was taken.

  The selection unit refers to the shooting time of the images included in the group, calculates a shooting interval between images, compares the shooting time with a predetermined threshold, and selects the template according to the comparison result. Can be.

  An image processing method according to one aspect of the present invention is a template for determining an image arrangement, and includes a management step for managing a template prepared for each number of images to be arranged, and an image attached to the image to be grouped. An acquisition step for acquiring information to be performed, a grouping step for dividing the image into one or more groups based on the information acquired in the processing of the acquisition step, and the group generated by the processing of the grouping step. A selection step of selecting the template managed in the process of the management step according to the number of images included, and an arrangement step of arranging the image in the template selected in the process of the selection step are included.

  A program according to one aspect of the present invention is a template for determining the arrangement of images, a management step for managing templates prepared for each number of images to be arranged, and information associated with the images to be grouped Included in the group generated in the grouping step and the grouping step of dividing the image into one or more groups based on the information acquired in the process of the acquiring step A computer including a selection step of selecting the template managed in the process of the management step according to the number of images, and an arrangement step of arranging the image in the template selected in the process of the selection step To run.

  In the image processing apparatus, method, and program according to one aspect of the present invention, an image is a template for determining the arrangement of images, and a template prepared for each number of images to be arranged is managed and images to be grouped The information attached to is acquired, the images are divided into one or more groups based on the acquired information, the managed template is selected according to the number of images included in the generated group, and the selected template is selected The image is arranged.

  According to one aspect of the present invention, one or more images can be arranged and displayed based on a predetermined template. The display can be a display reminiscent of the date and time when the image was taken.

  Embodiments of the present invention will be described below. Correspondences between constituent elements of the present invention and the embodiments described in the specification or the drawings are exemplified as follows. This description is intended to confirm that the embodiments supporting the present invention are described in the specification or the drawings. Therefore, even if there is an embodiment which is described in the specification or the drawings but is not described here as an embodiment corresponding to the constituent elements of the present invention, that is not the case. It does not mean that the form does not correspond to the constituent requirements. Conversely, even if an embodiment is described here as corresponding to a configuration requirement, that means that the embodiment does not correspond to a configuration requirement other than the configuration requirement. It's not something to do.

  An image processing apparatus according to one aspect of the present invention (for example, the image processing apparatus 1 in FIG. 1) is a template that determines the arrangement of images, and a management unit that manages templates prepared for each number of images to be arranged ( For example, based on the template DB in FIG. 2, an acquisition unit (for example, clustering unit 91 in FIG. 2) that acquires information associated with the images to be grouped, and the information acquired by the acquisition unit. The grouping means (for example, the clustering unit 91 in FIG. 2) that divides the image into one or more groups and the number of images included in the group generated by the processing of the grouping means are managed by the management means. Selection means (for example, the template setting unit 72 in FIG. 2) for selecting the template that has been selected, and the previous selection by the selection means A template, and a placement means for placing the image (e.g., recording and playback control unit 52 of FIG. 2).

  Embodiments of the present invention will be described below with reference to the drawings.

[Configuration of image processing apparatus]
FIG. 1 is a diagram showing a configuration of an information processing apparatus according to an embodiment of the present invention. The information processing apparatus 1 is an HDD (Hard Disc Recorder) recorder or the like.

  The image processing apparatus 1 in FIG. 1 includes an AV (Audio Video) signal supplied from an external device (not shown), a video signal input as an analog signal (not shown) or an analog signal supplied from a line tuner (not shown), and Record or play audio signals. The image processing apparatus 1 outputs the reproduced video signal to a monitor 2 connected to the outside. Further, the image processing apparatus 1 outputs the reproduced audio signal to the speaker 3 connected to the outside.

  External devices that supply AV signals include, for example, external devices connected via i.LINK (trademark), digital BS (Broadcasting Satellite) tuner or digital CS (Communications Satellite) tuner, digital video camera, digital still camera, Or a scanner.

  A drive 4 is connected to the image processing apparatus 1 as necessary. The drive 4 includes a magnetic disk (including a flexible disk), an optical disk (including a CD-ROM (Compact Disc-Read Only Memory), a DVD (Digital Versatile Disc), a magneto-optical disk), or a semiconductor memory. A removable medium 5 is loaded as necessary to exchange data.

  1 includes a codec 11, a sampling rate converter 12, an analog input processing unit 13, an A / D converter 14, a switching unit 15, an NTSC (National Television System Committee) decoder 16, an encoder 17, and a recording / reproducing unit 18. The operation unit 19, the control unit 20, the decoder 21, the NTSC encoder 22, and the D / A converter 23 are included.

  The codec 11 decompresses the compression of the digital video signal out of the AV signal supplied from an external device (not shown), and supplies it to the NTSC decoder 12.

  The sampling rate converter 12 converts the sampling rate of the digital audio signal among AV signals supplied from an external device (not shown) to a different sampling rate. The sampling rate converter 12 supplies the converted digital audio signal to the switching unit 15.

  The analog input processing unit 13 is based on an analog signal (not shown) or an analog signal (not shown) supplied from the line tuner based on a signal supplied from the control unit 20 and indicating an operation from the user in the operation unit 19. Select one of the following. The analog input processing unit 13 supplies an analog video signal of the selected analog signals to the NTSC decoder 16. Further, the analog input processing unit 13 supplies an analog audio signal among the selected analog signals to the A / D converter 14.

  The A / D converter 14 A / D converts the analog audio signal supplied from the analog input processing unit 13. The A / D converter 14 supplies a digital audio signal as a result of A / D conversion to the switching unit 15.

  The switching unit 15 is supplied from the control unit 20 and is supplied from the digital audio signal supplied from the sampling rate converter 12 or the A / D converter 14 based on a signal indicating an operation from the user in the operation unit 19. Select one of the digital audio signals. The switching unit 15 supplies the selected digital audio signal to the NTSC decoder 16.

  The NTSC decoder 16 converts the digital video signal supplied from the codec 11 or the analog video signal input from the analog input processing unit 13 into an NTSC digital video signal. The NTSC decoder 16 synthesizes the converted NTSC digital video signal and the digital audio signal supplied from the switching unit 15.

  When recording a video signal and an audio signal input to the image processing apparatus 1, the NTSC decoder 16 supplies a digital AV signal that is a synthesized signal to the encoder 17. On the other hand, when the video signal and the audio signal input to the image processing apparatus 1 are reproduced without being recorded, the NTSC decoder 16 converts a digital AV signal, which is a synthesized signal, into an NTSC encoder 22 and a D / A which will be described later. This is supplied to the converter 23.

  The encoder 17 performs an encoding process on the digital AV signal supplied from the NTSC decoder 16 according to a predetermined method such as an MPEG (Moving Picture Experts Group) method. The encoder 17 supplies the compression-coded signal data, which is the result of the encoding process, to the recording / reproducing unit 18.

  The recording / reproducing unit 18 is composed of, for example, a hard disk (HDD) or an optical disk. The recording / reproducing unit 18 records the signal data supplied from the encoder 17. In addition, the recording / reproducing unit 18 supplies the recorded signal data to the decoder 21. That is, the recording / reproducing unit 18 reproduces the recorded signal data.

  The operation unit 19 includes, for example, various operation buttons, keys, or a touch panel. The user operates the operation unit 19 in order to input an instruction to the image processing apparatus 1. The operation unit 19 supplies a signal indicating the user operation to the control unit 20 according to the user operation.

  The control unit 20 includes a microprocessor, for example, and controls each unit in the image processing apparatus 1 based on a signal indicating a user operation supplied from the operation unit 19.

  The decoder 21 performs a decoding process on the signal data supplied from the recording / reproducing unit 18 according to a method corresponding to, for example, the MPEG method. The decoder 21 supplies the digital video signal of the decompressed and decoded digital AV signal as a result of the processing to the NTSC encoder 22. Further, the decoder 21 supplies a digital audio signal of the decompressed and decoded digital AV signal as a result of the processing to the D / A converter 23.

  The NTSC encoder 22 converts the digital video signal supplied from the decoder 21 into an NTSC video signal. The NTSC encoder 22 supplies the converted video signal to the monitor 2.

  The D / A converter 23 D / A converts the digital audio signal supplied from the decoder 21. The D / A converter 23 supplies an audio signal as a result of the conversion to the speaker 3.

  With such a configuration, the image processing apparatus 1 can record or reproduce the input video signal and audio signal. Further, the image processing apparatus 1 can reproduce the video signal and the audio signal recorded in the recording / reproducing unit 18.

  The image processing apparatus 1 has a function of inputting and recording or reproducing a still image signal (data) in addition to a video signal. That is, the image processing apparatus 1 according to the present embodiment is an apparatus capable of handling moving images and still images. For example, the encoder 17 and the decoder 21 are methods suitable for moving images and methods suitable for still images. Is configured to support.

  In the configuration example shown in FIG. 1, the configuration for processing the video signal is mainly illustrated. However, the same configuration may be used so that the image data of the still image can be processed. A configuration in which still image data is directly recorded in the reproduction unit 18 may be employed, and the configuration illustrated in FIG. 1 can be appropriately changed.

  FIG. 2 is a diagram illustrating a configuration example of the recording / reproducing unit 18 and the control unit 20 of the image processing apparatus 1 of FIG.

  The recording / playback unit 18 includes a still image database 31, a moving image database 32, a template database 33, and a content database 34. Hereinafter, the still image database 31 is described as a still image DB 31, the moving image database 32 is described as a moving image DB 32, the template database 33 is described as a template DB 33, and the content database 34 is described as a content DB 34.

  The still image DB 31 records still image data, which is still image data supplied from an external device, or still image data (thumbnail data of thumbnail images) of a still image extracted from a moving image as will be described later. . The moving image DB 32 stores moving image data which is moving image data supplied from the encoder 17 of FIG. In the following description, for example, “image recorded in the still image DB 31” is described, and this “image” is expressed including the meaning of “image data”.

  The template DB 33 is information for setting a background template that is data of a background image that is a background of an image arranged in the album and an arrangement of images in the album when an album that is content recorded in the content DB 34 is displayed. Store the placement template.

  The content DB 34 records album data. The details of the album will be described later. As will be described later, the album is composed of pages with still images pasted, and can be handled in the same way as an album on paper media. Sometimes it is done.

  The control unit 20 includes a selection unit 51, a recording / playback control unit 52, a display control unit 53, and a content creation unit 54.

  The selection unit 51 generates information indicating the content operated by the user based on the signal indicating the user operation supplied from the operation unit 19 and supplies the information to the recording / playback control unit 52.

  For example, the selection unit 51 generates and records information indicating the album selected by the user based on a signal supplied from the operation unit 19 and indicating a user operation for selecting an album displayed on the monitor 2. This is supplied to the reproduction control unit 52.

  Further, for example, the selection unit 51 receives information indicating a predetermined folder selected by the user based on a signal supplied from the operation unit 19 and indicating a user operation for selecting a predetermined folder in which an image is stored. It is generated and supplied to the recording / playback control unit 52.

  The recording / playback control unit 52 controls recording or playback performed by the recording / playback unit 18. The recording / playback control unit 52 reads image data, a background template, an arrangement template, or album data from the recording / playback unit 18.

  For example, the recording / playback control unit 52 receives image data stored in a predetermined folder from the still image DB 31 or the moving image DB 32 based on the information supplied from the selection unit 51 and indicating the predetermined folder selected by the user. read out.

  Further, the recording / playback control unit 52 supplies the read image data, background template, arrangement template, or album data to the content creation unit 54. The recording / playback control unit 52 supplies the content (album data) created by the content creation unit 54 to the content DB 34 for storage.

  The display control unit 53 controls the decoding process of the decoder 21 based on the image data, background template, arrangement template, or album data read by the recording / playback control unit 52.

  The content creation unit 54 creates content using the image data supplied from the recording / playback unit 18 under the control of the recording / playback control unit 52. The content creation unit 54 includes a calculation unit 71 and a template setting unit 72.

  The computing unit 71 divides a plurality of images into groups (clustering) based on the data accompanying the images supplied from the recording / playback control unit 52. The computing unit 71 includes a clustering unit 91 and a hierarchy determining unit 92.

  The clustering unit 91 performs image clustering based on data accompanying the image supplied from the recording / playback control unit 52. The hierarchy determining unit 92 determines a cluster hierarchy that is a clustering result of the clustering unit 91.

  The template setting unit 72 instructs the recording / reproducing unit 52 to read the background template and the arrangement template from the template DB 33, or sets the read background template and arrangement template to the generated cluster. To do.

  The content creation unit 54 records the album in which the background template and the arrangement template are set in the content DB 34 of the recording / playback unit 18 via the recording / playback control unit 52.

  The operation of the image processing apparatus 1 having such a configuration will be described. In the following description, the operation of the image processing apparatus 1 when creating an album will be described as an example.

[About creating an album (content)]
With reference to the flowchart of FIG. 3, an outline of processing related to album creation will be added, and then details of processing in each step will be described with reference to other drawings.

  In step S11, image data is read. The recording / playback control unit 52 reads image data from the still image DB 31. In step S11, not the image data main body but data associated with the image data required for the clustering process performed in step S12 is read. For example, an EXIF (Exchangeable Image File Format) tag is read as data accompanying the image data.

  The timing at which image data is read, that is, in this case, the timing at which an album is created includes the following timing.

  Timing 1 is when there is an instruction from the user. When the user wants to create an album, the user operates the operation unit 19 to issue an album creation instruction. Even when the user gives an instruction, for example, it may be created at the timing when the user gives an instruction to create an album. When the user gives an instruction to play an album, the album is created and played back. You may be made to do.

  Timing 2 is when a folder for managing still images is newly created. When a folder is created, an album is created for the images managed in that folder.

  Timing 3 is when a new image is added to an already created folder. When a new image is added to an existing folder, an album including the new image is created. In other words, it is re-edited into an album including a new image.

  Timings 1 to 3 are examples of timing and do not indicate limitations. Therefore, album creation (the process of step S11) may be started at a timing other than timings 1 to 3. Further, the timing is not limited to one, and timings 1 to 3 and other timings can be used in combination.

  In step S11, when information accompanying the image data and required for the clustering process is read, the clustering process is started in step S12. Details of the clustering process will be described later. By executing the clustering process, the plurality of images are divided into a plurality of groups. When attention is paid to one group, clustering is performed so that the images in the noticed group have some relationship. Some relevance is temporal relevance here, as will be described below.

  When clustering is performed in step S12, a background template is read out in step S13 based on the processing result. In step S14, the arrangement template is read out.

  Here, the background template and the arrangement template will be described.

  FIG. 4 is a diagram illustrating an example of the configuration of the background template. The background templates 111-1 to 111-3 shown in FIG. 4 have different pictures and accompanying information. In the following description, when it is not necessary to individually distinguish the background templates 111-1 to 111-3, they are simply described as the background template 111. The same description is made for an arrangement template to be described later.

  For example, the background templates 111-1 to 111-3 each have information about date and time as accompanying information, and when images are clustered according to the shooting date and time, information on the date and time of the clustered images and the background template 111. The background template 111 that matches the date and time information attached to -1 to 111-3 is selected and set by the template setting unit 72.

  For example, if the background template 111-1 has date / time information “January 1” and the shooting date / time of a predetermined image is around January 1, the background template 111-1 is set for the image. Is done.

  In addition, when the background template 111 has information about an event and the image is clustered by the event (for example, when clustered based on the event name (title) given by the user), the event of the clustered image The template setting unit 72 sets a background template in which the information matches the event information attached to the background template 111.

  For example, when the background template 111-1 has event information “New Year” and the event name shown in the predetermined image is given as New Year, the background template 111-1 is set in the image. The

  The background template 111 is not limited to the background templates 111-1 to 111-3 shown in FIG. 4, and three or more background templates are stored in the template DB 33. Is set. Further, for example, not only the background template 111 in which a picture is drawn as shown in FIG. 4, for example, the background template 111-1, but also the white background template 111 or the like is used as the background template 111. And is set as a setting target by the template setting unit 72.

  5 to 9 are diagrams showing examples of arrangement templates stored in the template DB 33. FIG. Each placement template is a template used when placing a predetermined number of images, and a plurality of templates are prepared for each number of placed images. Here, when a plurality of templates for arranging a predetermined number of images, for example, one image, are described together, they are described as an arrangement template group.

  Arrangement templates 131-1 to 131-3, arrangement templates 132-1 to 132-3, arrangement templates 133-1 to 133-3, and arrangement template 134-1 included in the arrangement template groups shown in FIGS. Through 134-3 and the placement templates 135-1 through 135-3 each have associated information. The accompanying information is, for example, information related to the date and time and the event, like the background template 111.

  The arrangement template group including the three templates of the arrangement templates 131-1 to 131-3 shown in FIG. 5 is the arrangement template 131 when one image is arranged on one page. The arrangement templates 131-1 to 131-3 are common in that one image is arranged, but the positions and sizes of the one image in the page are set differently. Template.

  The arrangement template group including the three templates of the arrangement templates 132-1 to 132-3 shown in FIG. 6 is the arrangement template 132 when two images are arranged on one page. The arrangement templates 132-1 to 132-3 are common in that two images are arranged, but the positions and sizes of the two images are different in the page. It is a template set as follows.

  The arrangement template group including the three templates of the arrangement templates 133-1 to 133-3 shown in FIG. 7 is the arrangement template 133 when three images are arranged on one page. The arrangement templates 133-1 to 133-3 are common in that three images are arranged, but the positions and sizes of the three images are different in the page. It is a template set as follows.

  The arrangement template group including the three templates of the arrangement templates 134-1 to 134-3 shown in FIG. 8 is the arrangement template 134 when four images are arranged on one page. The arrangement templates 134-1 to 134-3 are common in that four images are arranged, but the positions and sizes of the four images in the page are different. It is a template set as follows.

  The arrangement template group including the three templates of the arrangement templates 135-1 to 135-3 shown in FIG. 9 is the arrangement template 135 when five images are arranged on one page. The arrangement templates 135-1 to 135-3 are common in that five images are arranged, but the positions and sizes of the five images in the page are different. It is a template set as follows.

  In each of the placement templates 131 to 135, the position and area (size) where each image is displayed are fixed. For example, one image is displayed on the placement template 131-1, but the position and size at which the image is displayed are fixed. If the image has a size larger than the displayed size, Even if the image is arranged in the arrangement template 131-1, the size of the area in which the image is displayed is not changed. Therefore, when a large-size image is pasted on the placement template, a part of the image is cut out, pasted on the placement template, and displayed.

  The arrangement templates 131 to 135 shown in FIGS. 5 to 9 are examples and do not indicate limitations. In addition, here, an example is shown up to the case where a maximum of five images are displayed on one page, but an arrangement template in which five or more images are displayed on one page is stored in the template DB 33 and used. Is of course good. However, if the number of images displayed on one page is increased, the display area per sheet becomes smaller, which may make it difficult to see the image. Therefore, an arrangement template that arranges too many images on one page may be used. It is not preferable to prepare.

  The template setting unit 72 sets an appropriate arrangement template among the arrangement templates 131 to 135 shown in FIGS. 5 to 9 stored in the template DB 33. Although processing for setting an appropriate arrangement template will be described later, it is performed using the result of clustering.

  Returning to the description of the flowchart of FIG. 3, when a background template is set in step S13 and an arrangement template is set in step S14, an image pasting process is performed in step S15. Although detailed description of the image pasting process executed in step S15 will be described later, this is a process of pasting an image on a set arrangement template and combining it with a background template.

  In this way, a predetermined page in the album is created. The processing in steps S11 and S12 is performed on a predetermined folder, and as a result, a plurality of still images in one folder are classified into a plurality of clusters. Then, the processing in steps S13 to S15 is executed for one cluster, so that one page in the album is created.

  An example of one page in the created album is shown in FIG. In the example of FIG. 10, a background template 111-1 and an arrangement template 135-1 are set on the page 151. The background template 111-1 has, for example, the same cluster information as the cluster information of the five images arranged on the page 151. Further, for example, the arrangement of five images is determined by the arrangement template 135-1. In addition, a comment “◯ Δ □ × 2005” is displayed on the page 151. Thus, the user may be able to set an arbitrary comment at an arbitrary position on the page 151 by operating the operation unit 19.

  Thus, a predetermined page in the album is an image in which an image is pasted on the set background template based on the set layout template.

  Further, an image arranged in the arrangement template, in other words, an image displayed on a predetermined page in the album is a still image, but the still image is only a still image stored in the still image DB 31. Instead, still images extracted from moving images stored in the moving image DB 32 may be included.

  For example, when creating an index or the like of a moving image stored in the moving image DB 32, a still image indicating one scene of the moving image is provided to a page in the album. It may be displayed. Such a still image may be referred to as a thumbnail image. In addition, a plurality of thumbnail images may be created from one moving image, and all of the plurality of thumbnail images may be displayed on a page constituting the album or selected. Only one thumbnail image may be displayed.

  FIG. 11 shows a screen example of a page when a still image extracted from a moving image is displayed. In FIG. 11, among the three still images displayed on the page 151 ′, the still image 161 is a still image (thumbnail image) extracted from the moving image. As a display for allowing the user to recognize that the still image 161 is a still image extracted from a moving image, a mark 162 “Video” is displayed in the vicinity of the still image 161.

  As described above, when the still image 161 on which the mark 162 is displayed is selected by the user, the reproduction of the original moving image of the still image 161 is started. “Selected by the user” means that, for example, when a cursor (not shown) is positioned on the still image 161, a predetermined operation is performed such as a mouse (not shown) being double-clicked. means.

  When the still image 171 or the still image 172 displayed on the page 151 'is selected, the selected still image 171 (still image 172) is enlarged and displayed.

  Returning to the description of the flowchart of FIG. 3, in step S <b> 16, it is determined whether or not processing has been completed for all clusters. In other words, the process of step S16 is a process for determining whether all pages in the album have been created.

  If it is determined in step S16 that the process has not been completed for all clusters, the process is returned to step S13, and the subsequent processes are executed. That is, the next page is created.

  On the other hand, if it is determined in step S16 that the processing has been completed for all clusters, the process proceeds to step S17. The processing comes to step S17 when the creation of all the pages in the album is completed, in other words, when the album is completed. Therefore, in step S17, the created album is stored (saved) in the content DB 34.

  In this way, an album is created. FIG. 12 shows an example of the configuration of the created album.

  In the example shown in FIG. 12, the album 181 is composed of five pages 151 to 155. The images arranged on the pages 151 to 155 of the album 181 are images stored in the same folder. On pages 151 to 155, a predetermined number of images determined by the arrangement template are arranged.

  Next, the process of each step described with reference to the flowchart of FIG. 3 will be described in detail.

[About clustering processing]
First, the clustering process executed in step S12 will be described.

  First, the concept of clustering performed in the present embodiment will be described. Note that the clustering method described below is an example, and does not mean that the present invention is applied only to the clustering method described below.

  FIG. 13 is a diagram illustrating an example of image groups (clusters). In FIG. 13, the horizontal axis indicates the time axis.

  FIG. 13 shows an example in which the images p1 to p12 photographed at the respective timings from time t1 to t12 are targeted for clustering. For example, a still image recorded in the still image DB 31 and a still image extracted from the moving image recorded in the moving image DB 32 are targeted for clustering. One square in FIG. 13 represents one image.

  In the image processing apparatus 1, the time interval of the photographing time of each image is obtained, and one cluster is formed from continuous images whose time interval distribution satisfies a certain condition. Each photographed image is added with an exchangeable image file format (EXIF) tag as attribute information, and information on the photographing time included in the EXIF tag is used for clustering.

  In the example of FIG. 13, a condition A and a condition B are set as conditions for defining a cluster, and one of the images p1 to p12 is defined by the condition A. The event name “wedding” is set in the cluster defined by the condition A.

  Although details of clustering will be described later, in the cluster in which the event name “wedding” is set, for example, the degree of variation in the time interval of the shooting time of each of the images p1 to p12 is smaller than a certain threshold value. It is stipulated from things.

  In the example of FIG. 13, one cluster is defined from the images p1 to p3 among the images p1 to p12 and one cluster is defined from the images p4 to p7. One cluster is defined from the images p8 to p12.

  “Church Ceremony” is set for the cluster consisting of images p1 to p3, “Reception” is set for the cluster consisting of images p4 to p7, and “Secondary Party” is set for the cluster consisting of images p8 to p12. .

  In the cluster in which the event name “Church Ceremony” is set, the image p3 and the images p1 to p3 constituting the cluster have the same degree of variation in the time interval of the photographing time, whereas the image p3, Next (next on the time axis), the time interval with the image p4 which is the first image among the images p4 to p7 which is a group of images with a similar degree of variation in the time interval of the photographing time is relatively large. This is specified because it was determined that there was a change in the frequency of shooting.

  Also, in the cluster in which the event name “Reception Party” is set, the degree of variation in the time interval of the shooting time of each of the images p4 to p7 constituting the event is close, while the image p7 and the next In addition, the time interval with the image p8 which is the first image among the images p8 to p12, which is a group of images having a close variation in the time interval of the shooting time, is relatively large, and the frequency of shooting changes at that portion. It was specified because it was judged that there was.

  In the cluster in which the event name “secondary party” is set, the degree of variation in the time interval of the photographing time of each of the images p8 to p12 constituting the cluster is close, whereas the image p12 is photographed next. It is specified because the time interval with the first image of a group of images with similar time interval variations is relatively large, and it was determined that there was a change in the frequency of shooting at that part. is there.

  The event names of “wedding”, “ceremony at church”, “banquet”, and “second party” are set manually by the user, for example.

  Thus, in the image processing apparatus 1, a plurality of conditions are set as conditions for clustering the same target image, and clusters of different granularities are defined based on the respective conditions. In the example of FIG. 13, the condition A is a condition that defines a cluster with a low particle size (coarse), and the condition B is a condition that defines a cluster with a particle size higher than the condition A (fine).

  The images included in each cluster defined as described above are virtually managed together by one folder and presented to the user in a hierarchical structure.

  FIG. 14 is a diagram illustrating an example of a hierarchical structure.

  In the example of FIG. 14, a folder representing a cluster in which the event name of “wedding” defined by the condition A is set as a folder in a lower hierarchy of the folder in which the name of “my memory” is set; A folder representing a cluster with an event name of “Hokkaido Travel” is shown.

  In addition, the event names of “ceremony at church”, “banquet”, and “second party” are defined as the folders below the folder representing the cluster in which the event name of “wedding” is set. A folder representing a cluster in which is set is shown.

  Furthermore, as a folder in the lower hierarchy of the folder representing the cluster in which the event name “Hokkaido Travel” is set, “Kushiro Wetland”, “Sapporo Susukino”, “Wakkanai Crab”, “Furano” A folder representing a cluster in which each event name of “Abashiri Prison” is set is shown. The clusters represented by these folders are also defined by the distribution of the time intervals of the shooting times of the images constituting each cluster.

  As described above, in the image processing apparatus 1, a folder is created for each event, and an image taken at each event is included in the created folder. By selecting a predetermined folder from, you can view and organize images by event.

  For example, the user does not simply browse images obtained by shooting in chronological order as shown in the upper part of FIG. 15, but as shown at the end of the white arrows, the folders f11, f12, and f122. , F211, f212, and f22, that is, by selecting a folder with a desired granularity and browsing the images in the desired order.

  One ellipse in FIG. 15 represents one folder. In FIG. 15, the entire image obtained by photographing is divided into an image included in the folder f1 and an image included in the folder f2, and the image included in the folder f1 is further divided into an image and a folder included in the folder f11. It is divided into images included in f12. An image included in the folder f11 is divided into an image included in the folder f111 and an image included in the folder f112, and an image included in the folder f12 is divided into an image included in the folder f121 and an image included in the folder f122. Similarly, the images included in the folder f2 are also divided into images included in lower-level folders.

  FIG. 16 is necessary for executing the clustering process while extracting the portion for executing the clustering process in the image processing apparatus 1 from the configuration of the image processing apparatus 1 shown in FIG. 1 or 2. It is the figure which added the part.

  The parts that execute the clustering process shown in FIG. 16 are a still image DB 31, a moving image DB 32, a calculation unit 71 (clustering unit 91, hierarchy determination unit 92), a tag reading unit 201, a condition setting unit 202, and an event name. A setting unit 203 is included.

  In the still image DB 31, an EXIF tag including information such as a shooting time and a shooting date is recorded in association with the shot image.

  When performing image clustering, the tag reading unit 201 reads the shooting time of each image to be clustered from the EXIF tag recorded in the still image DB 31, and outputs the read shooting time to the calculation unit 71. . Further, as will be described later, the tag reading unit 201 reads information related to the shooting time of the moving image recorded in the moving image DB 33 when there is an instruction from the user, and the read shooting time is sent to the calculation unit 71. Output.

  Image clustering may be performed each time a single image is taken, or may be performed at a predetermined timing such as when the user instructs clustering. In the present embodiment, it is assumed that it is performed at the timing when the album is created (assumed to be performed as the process of step S12).

  The computing unit 71 includes a clustering unit 91 and a hierarchy determining unit 92. The clustering unit 91 acquires cluster data recorded in the content DB 34, and performs clustering of images based on the acquired cluster data and the shooting time supplied from the tag reading unit 201. For example, the clustering unit 91 obtains a clustering result in which the entire image including a clustered image in the past is managed in a form having a binary tree structure.

  The cluster data recorded in the content DB 34 includes information that represents the cluster to which each image belongs and information that represents the binary tree structure of the entire image.

  The hierarchy determining unit 92 determines a cluster that meets the condition set by the condition setting unit 202 based on information representing the binary tree structure that is the clustering result acquired by the clustering unit 91. For example, information indicating which image each cluster includes or information indicating which cluster each image belongs to is acquired by the hierarchy determination unit 92 to represent the binary tree structure of the entire image, and finally Is recorded in the content DB 34 as cluster data representing the result of a typical clustering.

  The condition setting unit 202 sets a grouping condition that is a condition for defining a cluster and outputs the grouping condition to the hierarchy determination unit 92. The grouping condition may be set by being selected by a user from a plurality of conditions prepared in advance, or a plurality of clusters may be combined into one cluster, If the history of cluster editing performed by the user, such as dividing into clusters, is managed, the granularity of the cluster considered to be preferred by the user is obtained by learning using the editing history. Conditions that can obtain a cluster of granularity may be dynamically set.

  The content DB 34 records the cluster data supplied from the hierarchy determination unit 92 in association with the image data recorded in the still image DB 31. The event name supplied from the event name setting unit 203 is also recorded in association with each cluster represented by the clustering result. As described above, in the image processing apparatus 1, the image data main body and the cluster data are managed separately.

  The event name setting unit 203 outputs the event name set by the user to the content DB 34 and records it in association with the cluster. Further, when the event name is not set by the user, the event name setting unit 203 identifies the images belonging to each cluster using the cluster data recorded in the content DB 34, and the shooting date and the shooting time zone of the identified image. Event names including (AM, PM, etc.) are recorded in association with clusters.

  Here, details of processing performed by the calculation unit 71 will be described.

  For example, with 100 images p1 to p100 as clustering targets, these images are classified into event cluster A (cluster of images satisfying grouping condition A) and sub-event cluster B (cluster of images satisfying grouping condition B). The process to divide will be described.

  First, the creation of a binary tree structure performed by the clustering unit 91 will be described with reference to FIGS.

  In FIG. 17 to FIG. 25, one circle in which letters “p” and numbers are written represents one image, and one circle in which letters “n” and numbers are written. A circle represents one node. Also, the right side of the figure is assumed to be a time later than the left side. Here, a case will be described in which images p1 to p100 are clustered in order of photographing time (image p1 is the oldest data and image p100 is the newest data). After the images p1 and p2 are targeted for clustering (after being incorporated into the binary tree structure), the image p3 that is newer than the image p1 and older than the image p2 is newly targeted for clustering, not in order of shooting time. The clustering performed for each image will be described later.

  When clustering has not yet been performed and the image p1 obtained by the first shooting is inserted into the tree (when it is targeted for clustering), there is no node that is the root of the tree in the initial state. As shown in FIG. 17, the image p1 itself becomes a root node.

  When an image p2 photographed subsequent to the image p1 is inserted into the tree, a node n1 is newly created as shown in FIG. 18, and the photographing time of the image p2 is later than the photographing time of the image p1. , Node n1 is connected with image p1 as a left child node and image p2 is connected as a right child node. The node n1 becomes a root node instead of the image p1.

  The shooting time of the image p1 is recorded as the minimum time of the node n1, and the shooting time of the image p2 is recorded as the maximum time. As the time of a node, an average value (intermediate time) of photographing times of two child nodes having the node as a parent node may be recorded.

The image p3 taken _subsequent to the image p2 is inserted into the tree, and as shown in FIG. 19, the time interval td _p2p3 between the shooting time of the image p3 and the shooting time of the image p2 is the shooting time of the image p2 and the image p1. 20 is smaller than the time interval td _p1p2 , the node n2 is newly created as shown in FIG. 20, and the image p2 is connected to the node n2 as its left child node, and as the right child node. The image p3 is connected. Further, the node n2 is connected to the node n1 as a right child node instead of the image p2.

  The shooting time of the image p2 is recorded as the minimum time of the node n2, and the shooting time of the image p3 is recorded as the maximum time. At this time, the maximum time of the node n1, which is the parent node of the node n2, is overwritten with the shooting time of the image p3.

An image p4 taken after the image p3 is inserted into the tree, and as shown in FIG. 21, the time interval td _p3p4 between the shooting time of the image p4 and the shooting time of the image p3 is the shooting time of the image p3 and the image p2. greater than the time interval td _p2p3 shooting time, and the maximum time interval td _N2maxp4 the photographing time and the node n2 of the image p4 is, when the time intervals greater than td _P1n2min the capture time of the minimum time and the image p1 of the node n2 As shown in FIG. 22, a node n3 is newly created as a root node. Further, the node n3 is connected to the node n1 as the left child node, and the image p4 is connected to the node n3 as the right child node.

  The minimum time of the node n1 is recorded as the minimum time of the node n3, and the shooting time of the image p4 is recorded as the maximum time.

An image p5 taken _subsequent to the image p4 is inserted into the tree, and as shown in FIG. 23, the shooting time of the image p4 and the node n1 from the time interval td _p4p5 between the shooting time of the image p5 and the shooting time of the image p4. When the time interval td _n1maxp4 of the maximum time is larger, a node n4 is newly created as shown in FIG. Further, the node n4 is connected with the image p4 as a left child node thereof and with the image p5 as a right child node. Further, the node n4 is connected to the node n3 as a right child node instead of the image p4.

  The shooting time of the image p4 is recorded as the minimum time of the node n4, and the shooting time of the image p5 is recorded as the maximum time. At this time, the maximum time of the node n3 that is the parent node of the node n4 is overwritten with the shooting time of the image p5.

  At this point, the binary tree structure is formed from the nodes n1 to n4 including the five images p1 to p5.

  Since the image included in each node is an image that is directly connected to each node or indirectly connected to another node, when the tree as shown in FIG. The images included in n3 are five images p1 to p5, and the images included in the node n1 are three images p1 to p3. The image included in the node n2 is two images p2 and p3, and the image included in the node n4 is two images p4 and p5.

  In this way, each time a new image is inserted, images having a smaller time interval between shooting times or images and nodes having a smaller time interval between the shooting time and the set time are hung on the same node. It will be connected to.

When shooting is repeated and images p6 to p100 are inserted into the tree, the same processing is performed, and finally the images p1 to p100 are included in the root node n _root as shown in FIG. A binary tree structure is obtained.

  Next, clustering based on the grouping condition performed by the hierarchy determination unit 92 will be described with reference to FIGS. 26 to 32.

  In the hierarchy determination unit 92, attention is paid to each node in the binary tree structure, and the standard deviation sd of the shooting time intervals of all the images in the node is calculated by the following equation (1).

N is the number of time intervals of image capturing time, and is represented by “the number of images included in a node−1”. td _n is the _nth time interval of N time intervals as viewed from the previous time point. Td with “￣” is an average value of the time interval td in the node.

  Further, the deviation dev (the absolute value of the difference between the time interval between the child nodes and the average of the time interval of the photographing time) between the child nodes having the node of interest as the parent node is expressed by the following equation (2). Calculated.

N is the number of time intervals of image capturing time, and is represented by “the number of images included in a node−1”. td _c is the time interval between child nodes. Td with “￣” is an average value of the time interval td in the node. Note that the time interval between child nodes is the time at which the last image in time included in the child node that is the parent of the two child nodes having the node of interest as the parent node is temporally The time interval of the shooting time of the first image included in the subsequent child node. Specific examples will be described later.

  Further, the ratio of the deviation dev calculated by the equation (2) to the standard deviation sd calculated by the equation (1) is set as the division parameter th at the node of interest. The division parameter th is expressed by the following equation (3), and is a parameter serving as a reference for determining whether or not to divide a child node whose parent node is the node of interest as belonging to different clusters.

  Here, the value obtained by the above equation will be specifically described with reference to FIG.

  FIG. 26 is a diagram showing a part of the entire binary tree structure created by the clustering unit 91 in the same range as FIG.

In FIG. 26, td ₁ is the time interval between the shooting time of the image p1 and the shooting time of the image p2, and td ₂ is the time interval between the shooting time of the image p2 and the shooting time of the image p3. Td ₃ is a time interval between the shooting time of the image p3 and the shooting time of the image p4, and td ₄ is a time interval between the shooting time of the image p4 and the shooting time of the image p5.

  For example, when attention is paid to the node n3 that is the node closest to the root node in FIG. 26, first, each value is substituted into the equation (1), and the standard deviation sd is expressed by the following equation (4). The Moreover, the average value of the time interval of imaging time is represented by the following formula (5).

  The deviation dev is expressed by the following equation (6).

  That is, two child nodes having the target node n3 as a parent node are the nodes n1 and n4, and the time of the last image p3 included in the previous node n1 in time and the time The time interval td3 of the photographing time of the first image 4 included in the subsequent node n4 is the time interval between the node n1 and the node n4, which is used to calculate the deviation dev when attention is paid to the node n3. It is done.

  From the standard deviation sd calculated by the equation (4) and the deviation dev calculated by the equation (6), the division parameter th when attention is paid to the node n3 is calculated and set to the node n3.

  When the setting of the division parameters for all the nodes is completed as described above, the hierarchy determining unit 92 next sets a threshold as a grouping condition.

  For example, when the threshold value a is set as the grouping condition A and the threshold value b is set as the grouping condition B, “a> b” such as “a = 3.0”, “b = 2.6”, etc. 27, as shown in FIG. 27, “cluster defined by grouping condition A> cluster defined by grouping condition B” (in the cluster defined by grouping condition A) Thus, several clusters defined by the grouping condition B can be created).

In the example of FIG. 27, the entire target image is divided into two clusters of groups g ₁ and g _{2 according} to the grouping condition A, and is divided into five clusters of groups g _{3 to} g _{7 according} to the grouping condition B. It has been.

  That is, as will be described later, when the value of the set partition parameter is larger than the threshold, the portion of the node for which the partition parameter is set becomes the boundary portion of the cluster. The portion is unlikely to become a boundary portion of the cluster, and therefore the cluster granularity becomes coarse when viewed as a whole. On the other hand, the smaller the threshold value, the more likely the node part becomes the boundary part of the cluster, and therefore the granularity of the cluster becomes finer when viewed as a whole.

  When the time interval between the shooting time of an image and the shooting time of the next image is 30 minutes or more (time interval td> 30 min) or more than 1 day (time interval td> 1 day), the time A grouping condition may be set so that a part having an interval becomes a boundary part of the cluster, that is, a grouping condition that defines an upper limit of the time interval of photographing times of images included in one cluster. Good. As a result, images having a time interval of the photographing time of 30 minutes or more than one day are included in different clusters.

  After the threshold value as the grouping condition is set, the hierarchy determination unit 92 next assigns each node based on the set threshold value and the division parameter th set for each node as described above. A split flag is set.

  For example, a group A division flag having a value of 1 is set for a node whose division parameter th exceeds the threshold a set as the grouping condition A, and a group A division flag having a value of 0 is set for a node that does not exceed the threshold a. Is set.

  When a value of 1 is set as the division flag, the time interval between the child nodes whose parent node is the node for which the division flag is set is an image of the entire image in the node of interest. Indicates that it is relatively larger than the time interval. On the other hand, when the value of 0 is set as the division flag, the time interval between the child nodes whose parent node is the node for which the division flag is set is the entire image in the node of interest. This means that it is not much different from the time interval of the shooting time.

  When setting of the value of the group A split flag is completed, the nodes in the binary tree structure are noticed in ascending order (in order from the leaves to the roots), and the time interval of the child nodes is the time of the imaging time of the entire image in the node The image is segmented with a node that is relatively larger than the interval, that is, a node for which a value of 1 is set as the group A division flag as a boundary, and a cluster is defined by the grouping condition A.

  FIG. 28 is a diagram showing an example of setting the group A division flag.

  In the example of FIG. 28, nodes n11 and n12 are connected as left and right child nodes to node n10 which is the node closest to the root node, and nodes n13 and n14 are respectively connected as left and right child nodes to node n11. Connected. Nodes n12 and n16 are connected to the node n12 as left and right child nodes, respectively, and nodes n17 and n18 are connected to the node n14 as left and right child nodes, respectively.

  In the example of FIG. 28, a value of 1 is set as the group A division flag in each of the nodes n10, n11, n12, and n14 out of these nodes, and the bold line The cluster is divided as shown in FIG.

  As shown on the right side of FIG. 28, the node n13 is a node including images p1 to p17, and the node n17 is a node including images p18 to p30. The node n18 is a node including images p31 to p68, and the node n15 is a node including images p69 to p93. Node n16 is a node including images p94 to p100.

  Therefore, when a cluster is divided with a node portion having a group A division flag value of 1 as a boundary, the images p1 to p100 that are currently targeted for clustering are as shown in FIG. Cluster (event cluster A).

That is, the cluster A ₁ is configured from the images p1 to p17 included in the node n13, and the cluster A ₂ is configured from the images p18 to p30 included in the node n17.

Further, a cluster A ₃ is composed of the images p31 to p68 included in the node n18, and a cluster A ₄ is composed of the images p69 to p93 included in the node n15. Further, a cluster A ₅ is composed of the images p94 to p100 included in the node n16.

  As described above, clustering based on the grouping condition A is performed, information indicating which range of images each cluster defined by the grouping condition A includes, or which cluster each image belongs to Is obtained as a clustering result based on the grouping condition A.

  When clustering based on the grouping condition A is performed, grouping based on the grouping condition B is performed in the same manner.

  That is, in the hierarchy determination unit 92, a group B division flag having a value of 1 is set for nodes whose division parameter th exceeds the threshold value b set as the grouping condition B, and 0 for nodes that do not exceed the threshold value b. A group B division flag having a value of is set.

  When the setting of the value of the group B split flag is finished, the nodes in the binary tree structure are noticed in ascending order, and the time interval of the child nodes is relatively larger than the time interval of the photographing time of the entire image in the node, An image is divided with a node having a value of 1 set as the group B division flag as a boundary, and a cluster is defined by the grouping condition B.

  FIG. 30 is a diagram showing an example of setting the group B division flag.

  In the example of FIG. 30, the node n10, which is the node closest to the root node, is connected to the node n11 as the left child node, and the nodes n11 are connected to the nodes n13 and n14 as the left and right child nodes, respectively. . Further, nodes n17 and n18 are connected to the node n14 as left and right child nodes, respectively, and nodes n19 and n20 are connected to the node n17 as left and right child nodes, respectively. Further, nodes n21 and n22 are connected to the node n18 as left and right child nodes, respectively, and nodes n23 and n24 are connected to the node n19 as left and right child nodes, respectively.

  In the example of FIG. 30, a value of 1 is set as the group B division flag in each of the nodes n10, n11, n14, n17, n18, and n19 among these nodes. As a boundary, clusters are divided as shown by bold lines.

  As shown on the right side of FIG. 30, node n13 is a node including images p1 to p17, and node n23 is a node including images p18 to p21. The node n24 is a node including images p22 to p26, and the node n20 is a node including images p27 to p30. The node n21 is a node including images p31 to p57, and the node n22 is a node including images p58 to p68.

  Therefore, when a cluster is divided with the boundary of a node portion where 1 is set as the value of the group B division flag, the images p1 to p68 among the images p1 to p100 that are currently targeted for clustering are shown in FIG. It is divided into respective clusters (sub-event cluster B) as shown in FIG.

That is, a cluster B ₁ is configured from the images p1 to p17 included in the node n13, and a cluster B ₂ is configured from the images p18 to p21 included in the node n23.

Further, a cluster B ₃ is configured from the images p22 to p26 included in the node n24, and a cluster B ₄ is configured from the images p27 to p30 included in the node n20. Further, a cluster B ₅ is configured from the images p31 to p57 included in the node n21, and a cluster B ₆ is configured from the images p58 to p68 included in the node n22.

  Clustering based on the grouping condition B is performed as described above, information indicating which range of images each cluster defined by the grouping condition B includes, or which cluster each image belongs to Is obtained as a clustering result based on the grouping condition B. The acquired clustering result is output to the content DB 34 and recorded.

  FIG. 32 is a diagram showing the clustering result based on the grouping condition A (FIG. 29) and the clustering result based on the grouping condition B (FIG. 31).

  As shown in FIG. 32, when grouping is performed based on a plurality of grouping conditions, clusters of clustering results have a nested relationship.

In the example of FIG. 32, the cluster A ₁ and the cluster B ₁ are clusters including the same range of images, and the cluster A ₂ includes clusters B _{2 to} B ₄ . Cluster A ₃ includes clusters B ₅ and B ₆ .

When the respective clusters have the relation as shown in FIG. 32, for example, as a folder of a lower level, representing the cluster A _2, folders representing clusters B ₂ through B ₄ are respectively displayed, the cluster A ₃ Folders representing clusters B ₅ and B ₆ are respectively displayed as folders below the folder to be represented.

  As described above, the computing unit 71 performs hierarchical clustering and clustering based on dispersion (clustering based on the average of the shooting intervals of the entire image). As a result, clustering is not performed so that images with close imaging intervals are included in the same cluster, but clustering is performed so that images with uniform time intervals are included in the same cluster.

  Therefore, it is possible to perform clustering in a form that is in line with the user's individual subjectivity, as compared with the case where clustering is performed so that images with close photographing intervals are included in the same cluster.

  In order to identify the event unit desired by the user and create a cluster for each event, it is necessary to pay attention to the frequency of shooting, not the shooting interval itself, and the part where the frequency has changed is used as the cluster boundary. Thus, the obtained cluster is closer to the unit of the event. In order to know the frequency of shooting, it is necessary to statistically analyze the entire image taken up to that point (the entire image to be clustered). Clustering based on is adopted.

  In addition, since each image is managed in a hierarchical structure, by changing the threshold value as a grouping condition, a plurality of clusters can be combined into a single cluster, or a single cluster can be combined into a plurality of clusters. The user can easily reconfigure the cluster, for example, by dividing the cluster. As described above, by increasing the threshold value, the granularity of the cluster can be coarsened, that is, a plurality of clusters can be combined into one cluster, and conversely, by decreasing the threshold value, the granularity of the cluster can be reduced. Refinement, that is, one cluster can be divided into a plurality of clusters.

  As a result, for example, when viewing images by slide show playback, you can jump in units of events (units of event clusters), as in the case of chapter jumps in video, or divide the event into sub-event units. It is possible to jump in (sub-event cluster unit).

  Furthermore, by performing clustering based on the average of the shooting intervals of the entire image, it is possible to set grouping conditions without considering individual differences in shooting styles. In other words, it is not necessary for the user to set individual grouping conditions, even if the user is a person with a shooting style that is not frequently photographed or a person with a shooting style that is frequently photographed.

  For example, if images taken in less than 30 minutes are to be images of the same cluster, if there is an interval of 30 minutes or more after the previous image is taken, the first taken image is taken next. Since each image is an image in a different cluster, when the user is a person with a shooting style that does not shoot frequently (when the user is a person who takes a picture with an interval of 30 minutes or more), One image at a time is taken as another cluster image.

  On the other hand, if you set the time interval, which is the standard for clustering, so that images taken during the day will be the same cluster image, it will be taken in a completely different event in the morning and afternoon. The captured images are all regarded as images of the same cluster. Since the clustering is performed based on the average of the shooting intervals of the entire image, the clustering can be performed according to the shooting style of the individual user.

  Next, processing of the image processing apparatus 1 shown in FIG. 16 for generating cluster data will be described with reference to the flowchart of FIG. That is, the details of the clustering process executed in step S12 of FIG. 3 will be described.

  In step 11 (FIG. 3), the tag reading unit 201 reads an EXIF tag associated with a still image in a folder that is a clustering target of the still image DB 31. If necessary, information related to the shooting date and time of the moving image stored in the moving image DB 32 is also read.

  In step S <b> 31, the tag reading unit 201 acquires the shooting time from the EXIF tag of the shot image, and outputs the acquired shooting time to the calculation unit 71. That is, this example shows an example in which clustering is performed every time one image is captured.

  In step S <b> 32, the clustering unit 91 of the calculation unit 71 is based on the shooting time supplied from the tag reading unit 201 and the cluster data recorded in the content DB 34, and the binary tree structure including the entire image to be clustered. Is created as described with reference to FIGS.

  In step S33, the hierarchy determination unit 92 pays attention to one node in a predetermined hierarchy of the binary tree structure, and proceeds to step S34, and sets the standard deviation sd of the time intervals of the shooting times of all images in the node of interest. calculate.

  In step S35, the hierarchy determining unit 92 calculates a deviation dev of the time interval between the child nodes whose parent node is the node of interest.

  In step S36, the hierarchy determining unit 92 calculates the ratio of the deviation dev calculated in step S35 to the standard deviation sd calculated in step S34, and records the calculated ratio as the division parameter th in the node currently focused on. To do.

  In step S37, the hierarchy determining unit 92 determines whether or not all nodes in the binary tree structure have been noticed. If it is determined that attention has not been given, the hierarchy determination unit 92 returns to step S33, switches the noticed node, and thereafter. Repeat the process.

  On the other hand, if the hierarchy determining unit 92 determines in step S37 that all nodes in the binary tree configuration have been focused, the hierarchy determining unit 92 proceeds to step S38 and sets a threshold as a grouping condition. When obtaining clusters of different granularities for the same image, a plurality of grouping conditions are set here.

  In step S39, the hierarchy determining unit 92 compares the threshold value as the grouping condition set in step S38 with the division parameter th set for each node. For example, the hierarchy determination unit 92 sets 1 for a node for which the division parameter th exceeds the threshold value. A split flag with a value of 0 is set, and a split flag with a value of 0 is set for nodes that do not exceed the threshold. When a plurality of grouping conditions are set, division parameters are set for each grouping condition.

  In step S40, the hierarchy determining unit 92 pays attention to each node of the binary tree structure in ascending order, divides the cluster with the node having a value of 1 set in the division flag as a boundary, and represents the range of each cluster Create data.

  In step S41, the hierarchy determining unit 92 stores the created cluster data in the content DB 34, and ends the process.

  In this way, clustering is performed.

  Such clustering is performed so as to obtain a final result depending on an arrangement template that arranges the maximum number of arrangement templates prepared as arrangement templates. That is, in this case, since the arrangement template for arranging the five images as shown in FIG. 9 is the arrangement template for arranging the maximum number of sheets, the number of images included in one group as a result of clustering. The clustering process described above is performed so that the number of images becomes five or less.

[Processing related to placement template selection and settings]
Next, processing when an arrangement template is selected based on the result of clustering will be described. That is, the details of the background template setting process executed in step S14 of FIG. 3 will be described.

  As described with reference to FIGS. 5 to 9, the background template is a template used when 1 to 5 images are arranged at a predetermined size in a predetermined position in one page. .

  FIG. 34 is a flowchart illustrating details of the background template setting process executed in step S14. In step S61, the template setting unit 72 (FIG. 2) sets an arrangement template group according to the number of clustered still images.

  For example, it is assumed that the grouping as shown in FIG. 35 is performed due to the execution of the clustering process. Group 1 includes three images taken on March 1, Group 2 includes two images taken on March 2, and Group 3 includes March 3 3 images taken on the day are included, group 4 contains 2 images taken on 3 March, and group 5 contains 5 images taken on 4 March An image is included, and group 6 includes one image taken on March 4 and four images taken on March 5.

  According to the above-described clustering technique, for example, the number of images taken on March 3 is five, but is divided into group 3 and group 4. In this way, even image groups shot on the same day can be divided into different groups when the shot times are separated. If the time taken is different, there is a high possibility that the image was taken in a different scene (different event), and there is a high possibility that the same event is a different scene.

  Group 6 includes one image shot on March 4 and four images shot on March 5. In this way, even when the dates are different, they are classified into the same group when it is determined that there is a time series. When there is a series of shooting times, there is a high possibility that the images are shot in the same scene (same event, same scene).

  Images classified into the same group are displayed on the same page. Images displayed on the same page should have relevance and series. As described above, since images that are considered to be related are classified into the same group, images displayed on the same page are highly related images.

  The example shown in FIG. 35 is a case where images taken from March 1st to March 5th are managed in a predetermined folder, so images included in one group are relatively short. The image was taken within the time. For example, if a predetermined folder contains one photo taken every month and 12 photos are included, and if clustering is performed using the above-described clustering method, the conditions for clustering are as follows. However, four groups are created and classified so that four images are included in one group. That is, images are classified for each season.

  As described above, the photographing time of images included in one group may be within a relatively short time or may be within a long time. However, the images in one group have some relationship and series.

  As described above, the images classified into the groups after the clustering are highly related so that they are displayed on the same page. In order to display on the same page, an arrangement template group is selected depending on the number of images managed in one predetermined group.

  For example, referring to FIG. 35, since group 1 includes three images, the arrangement templates set for group 1 are arrangement templates 133-1 to 133- in which three images are arranged. 3 (FIG. 7), one of the placement templates 133 is set.

  Further, for example, since group 5 includes five images, the arrangement template group set for group 5 is the arrangement templates 135-1 to 135-3 (FIG. 9) in which five images are arranged. Any one of the placement templates 135 is set.

  In this way, an arrangement template group is selected depending on the number of images included in the group to be processed, and one arrangement template is further selected and set from the arrangement template group.

  Referring to FIG. 9 again, three placement templates 135-1 to 135-3 are prepared as placement templates for placing five images. When five images are included in the group to be processed, one arrangement template is selected from these arrangement templates 135-1 to 135-3. The selection is performed according to a predetermined condition. Is called.

  For example, it may be selected at random. Even when selected at random, if the same arrangement template is controlled so as not to be continuously selected, it is possible to provide the user with a variety of albums.

  For example, you may make it select in order. For example, after the arrangement template 135-1 is selected, the arrangement template 135-2 may be selected, and the selection may be made according to a predetermined order.

  Further, as described below, a template may be selected based on the shooting time of images in a group. Another process of setting the placement template executed in step S14 (FIG. 3) will be described with reference to the flowchart of FIG.

  In step S71, the template setting unit 72 sets an arrangement template according to the number of still images clustered and the shooting interval.

  By performing the clustering process, for example, as shown in FIG. 35, the data is classified into a plurality of groups, and one or more images are included in each group. When the time when the images in the group are taken is compared, as described above, the images are taken within a series of times.

  For example, referring again to FIG. 35, group 1 includes three images. The shooting times of the three images are “0:10” and “0:12” on March 1, respectively. ", And" 0:13 ". Group 3 also includes three images. The shooting times of the three images are “3:03”, “3:43”, and “4:03” on March 3, respectively. It is.

  The three images included in group 1 are images captured during 3 minutes, while the three images included in group 3 are images captured during one hour. Thus, even if the number of images included in one group is the same, the difference value between the oldest shooting time and the newest shooting time is different. In other words, even when the number of images included in one group is the same, the time interval (imaging interval) at which images included in the group are captured is different. Therefore, an arrangement template is selected according to the shooting interval between images.

  Here, when the shooting interval between images is short, an arrangement template for image arrangement is selected so that the user can recognize it (intuitively understand), and the shooting interval between images is long. In some cases, an arrangement template on which an image is arranged is selected so that the user can recognize it (intuitively understandable).

  Such selection will be described with reference to FIGS. 37 and 38 with specific examples. FIG. 37 shows an arrangement template when three images are arranged, and FIG. 38 shows an arrangement template when five images are arranged. The arrangement template for the three images shown in FIG. 37 shows an example different from the background template 133 for the three images shown in FIG. 7, and the arrangement template for the five images shown in FIG. Shows an example different from the background template 135 for five images shown in FIG.

  The arrangement template 231 shown in FIG. 37 is an arrangement template set when the shooting interval is shorter than the arrangement template 232, and the arrangement template 232 is an arrangement template set when the shooting interval is shorter than the arrangement template 233. It is.

  That is, in this case, the arrangement template 231 is an arrangement template set when it is determined that the shooting interval is the shortest. As shown in FIG. 37, the arrangement template 231 is arranged such that images are arranged so that the images overlap each other in order to express that the shooting interval is short. In this way, by arranging the images themselves close to each other, it is possible to make a display that allows the user to recognize that the shooting interval is short.

  On the other hand, in this case, the arrangement template 233 is an arrangement template that is set when it is determined that the shooting interval is the longest. However, as shown in FIG. The images are arranged in such a manner that they do not overlap and are separated from each other and do not line up on a straight line. In this manner, by disposing the images themselves, it is possible to provide a display that allows the user to recognize that the shooting interval is long.

  The same applies to the placement templates 234, 235, and 236 on which the five images shown in FIG. 38 are placed. That is, in this case, the arrangement template 234 is an arrangement template set when it is determined that the shooting interval is the shortest, and the arrangement template 236 is an arrangement template set when it is determined that the shooting interval is the longest.

  As shown in FIG. 38, in the arrangement template 234, five images are arranged so as to overlap each other in order to express that the shooting interval is short, like the arrangement template 231 (FIG. 37). It is a template for making. Similarly to the arrangement template 234 (FIG. 37), the arrangement template 236 expresses that the shooting interval is long, so that the five images do not overlap each other and are separated at a straight line. It is a template for making it arrange | position without arranging in a line.

  As described above, the shooting interval between the images is expressed by the arrangement of the images, and the arrangement template is selected by the shooting interval.

  Again, a description will be given with reference to FIG. As shown in FIG. 37, a threshold A and a threshold B are provided to select one of the placement templates 231, 232, and 233, and the placement template is placed when the shooting interval is equal to or less than the threshold A. When the template 231 is selected and the shooting interval is greater than or equal to the threshold A and less than or equal to the threshold B, the placement template 232 is selected, and when the shooting interval is greater than or equal to the threshold B, the placement template 233 is selected.

  As described above, the arrangement template is selected by providing a threshold value related to the photographing time and comparing it with the threshold value.

  Similarly, a threshold value C and a threshold value D are provided to select one of the placement templates 234, 235, and 236 for placing the five images shown in FIG. In this case, the placement template 234 is selected when the shooting interval is less than or equal to the threshold C, the placement template 235 is selected when the shooting interval is greater than or equal to the threshold C and less than or equal to the threshold D, and is placed when the shooting interval is greater than or equal to the threshold D. A template 236 is selected.

  The threshold A, threshold B, threshold C, and threshold D are different from each other. Alternatively, the same threshold value may be the same value.

  For example, the threshold A and the threshold C are both threshold values used when determining whether or not to select the arrangement template 231 (234) when it is determined that the shooting interval is the shortest. Alternatively, since different arrangement templates are selected, different values may be used. This threshold value may be set to a value such that an appropriate arrangement template is appropriately selected at a design stage or the like or by user setting.

  In this way, the arrangement template is set by comparing the threshold and the shooting interval. The value of the shooting interval used when the threshold and the shooting interval are compared is calculated as the difference between the oldest shooting time and the latest shooting time among the shooting times of the images in the group to be processed. The difference value is used.

  Alternatively, when the shooting times of the images in the group to be processed are arranged in time order, the difference between adjacent times is calculated, the average value of the differences is further calculated, and the calculated average value is You may make it use as a value of imaging | photography space | interval.

  Alternatively, when the shooting times of the images in the group to be processed are arranged in time order, the difference between adjacent times is calculated, and the smallest value of the difference is set as the shooting interval value or the largest The value may be used as the value of the shooting interval.

  Of course, the photographing interval value to be compared with the threshold value may be determined by other methods.

  When attention is paid to one threshold value, the threshold value may be a fixed value or a variable value. For example, the threshold A may be a fixed value, and the same value may always be set. Alternatively, the threshold A may be a variable value, and may be a value that is set every time an image in one folder is processed.

  When the threshold value is a variable value, for example, the threshold value may be set when the images in one folder are classified into a plurality of groups as shown in FIG.

  When the threshold is fixed, for example, when images taken in one year are classified, the shooting time of images in each group is likely to be a relatively long time. Therefore, when the shooting interval is long There is a high possibility that the arrangement template 233 or the arrangement template 236 to be selected is selected.

  If the same arrangement template is selected continuously, there is a possibility that a variety of page configurations cannot be performed. In addition, if the same arrangement template is selected in succession, it is meaningless to prepare a plurality of arrangement templates, and the effect obtained by preparing a plurality of arrangement templates as described above cannot be obtained, and it is notable. .

  Therefore, in order to prevent such a situation, it is preferable that some processing is added according to the shooting interval of images in the folder to be processed even when the threshold value is set to a fixed value.

  When the threshold value is a variable value, the difference between the oldest shooting time and the latest shooting time is calculated for each of a plurality of classified groups, and the threshold value is determined by using the difference for each group. May be. In this way, for example, as shown in FIG. 35, when an image photographed during 5 days is classified, an image photographed for one year is different (not shown). A threshold value is used, and a more accurate arrangement template can be set.

  That is, in other words, by setting the threshold value to a variable value, the image group to be processed is an image group photographed in a short period of time and an image group photographed in a long period of time. A similar arrangement template can be set.

  As described above, the process of step S71 in the flowchart of FIG. 36 is performed by comparing the shooting interval with the threshold value. The details of the processing in step S71 have been described above, but a simple explanation will be given with reference to the flowchart of FIG. 39 again.

  In step S91, the number of images managed in the group to be processed is grasped, and an arrangement template group for arranging the number of images is selected. For example, when four images are included in the group to be processed, an arrangement template group including the arrangement templates 134-1 to 134-3 (FIG. 8) is selected.

  In step S92, the shooting interval of the group to be processed is calculated. As described above, a value calculated by calculating a difference, an average value, or the like is used for the shooting interval.

  In step S93, one arrangement template is selected from a plurality of arrangement templates included in the arrangement template group selected in step S91. As described above, the selection is performed by comparing the shooting interval and the threshold value. The selected placement template is set as a placement template for the group to be processed.

  As described above, the process in step S93 is performed by determining whether or not the shooting interval is equal to or greater than a set threshold value. When the threshold is a variable value, for example, a process for setting the threshold is executed as a process between step S92 and step S93.

  By performing such processing, for example, the arrangement template for each group shown in FIG. 35 is set as shown in FIG. In the example shown in FIG. 40, the arrangement templates 231 to 233 shown in FIG. 37 are prepared as the arrangement template group in which three images are arranged, and the arrangement template group in which five images are arranged is shown in FIG. The arrangement templates 234 to 236 shown in FIG. Illustration and description of the arrangement template group in which two images are arranged are omitted.

  As shown in FIG. 40, since group 1 includes three images, the arrangement template group shown in FIG. 37 is selected. And it is judged that the imaging | photography space | interval of the images in the group 1 is below the threshold value A, As a result, the arrangement | positioning template 231 is set.

  Since group 3 includes three images, the arrangement template group shown in FIG. 37 is selected. The shooting interval between the images in the group 3 is determined to be greater than or equal to the threshold A and less than or equal to the threshold B, and as a result, the arrangement template 232 is set.

  Since group 5 includes five images, the arrangement template group shown in FIG. 38 is selected. Then, it is determined that the shooting interval between the images in the group 4 is not less than the threshold C and not more than the threshold D, and as a result, the arrangement template 235 is set.

  Since group 6 includes five images, the arrangement template group shown in FIG. 38 is selected. Then, it is determined that the shooting interval between the images in the group 5 is equal to or less than the threshold value C, and as a result, the arrangement template 234 is set.

  Since the arrangement template is set in this manner, it is possible to prevent the user from being provided with a monotonous album in which the arrangement of the same image continues, and to provide an album with a variety of image arrangements. It becomes possible.

[About image pasting]
When the arrangement template is set in this way, the images in the group are pasted based on the arrangement template. That is, in step S15 in the flowchart of FIG. 3, an image pasting process is executed. The image pasting process executed in step S15 will be described.

  Here, the arrangement template will be described again. FIG. 41 shows an arrangement template 301 and an arrangement template 302. Both the placement template 301 and the placement template 302 are placement templates when two images are pasted. The difference between the arrangement template 301 and the arrangement template 302 is that the arrangement template 301 is an arrangement template in which images to be pasted are not overlapped, but the arrangement template 302 is an image to be pasted. The images are arranged so that a part of the images overlaps to form a placement template.

  In the arrangement template 301 and the arrangement template 302, an area where one image is displayed is described as a display area. In the arrangement template 301, the display area 311 and the display area 312 are both square. Images managed in the group (still images recorded in the still image DB 31) are also generally rectangular. Therefore, the shape of the display area 311 (312) and the shape of the displayed image (the shape of the frame of the image) are the same.

  In the arrangement template 302, the display area 321 is a rectangle, but the display area 322 is a hexagon. In this case, the shape of the display area 322 is different from the shape of the pasted still image. Even in such a case, the still image to be pasted is not pasted after being processed into a hexagon according to the shape of the display area 322, but a square still image is pasted into the display area 322. Displayed.

  A rectangle shown by a dotted line in FIG. 41 (a portion overlapping with the display area 322 is shown by a solid line) is an area where an image is pasted. Hereinafter, a region where an image is pasted is described as a pasting region.

  The display area 311 and the pasting area 411, the display area 312 and the pasting area 412, and the display area 321 and the pasting area 421 shown in FIG. 41 have the same shape and the same size. Therefore, since these areas overlap each other, only a solid line representing the display area is shown in FIG.

  However, in the display area 322 and the pasting area 422, the display area 322 is hexagonal as described above, whereas the pasting area 422 is square. The still image to be pasted on the arrangement template is cut out in accordance with the size of the pasting area (details such as the definition of this size will be described later, but the size is based on the aspect ratio).

  Further, the display area and the pasting area will be described with reference to FIG. The arrangement template 303 shown in FIG. 42 is provided with a heart-shaped display area 331 and an elliptical display area 332 as display areas. As described above, for example, even in the heart-shaped display area 311, the pasting area 431 is a rectangle. The displayed still image is not cut out according to the heart shape of the display area 331, but is cut out according to the shape of the pasting area 431.

  Similarly, even in the oval display area 332, the pasting area 432 is a quadrangle, and the cut out still image is a quadrangle having the same shape as the pasting area 432.

  Thus, for example, when the shapes are different as in the display area 331 and the pasting area 431, the image is cut out in the form of the pasting area 431, and the part other than the display area 331 is masked in the cut out image. Thus, the image is processed so as to be displayed only in the display area 331.

  Further, with reference to FIG. 43, when the display area 341 is an area provided obliquely on the background template 304, a rectangular area including the display area 341 is set as the pasting area 441. The The display area 341 is arranged obliquely with respect to the background template 304 but has a quadrangular shape. As described above, even when the display area 341 has a quadrangular shape, if the display area 341 is arranged at an angle, the image is cut out with the image frame of the pasting area 441 and the part other than the display area 341 is masked. As a result, the user side is provided with a page as if the image was pasted diagonally.

  As shown in FIG. 43, when an image is displayed on the display area 341 arranged obliquely, the same pasting area 441 as the display area 341 may be provided. If the display area 341 and the pasting area 441 are the same, it is necessary to convert the pasted image into an oblique image. In order to perform image conversion, a large amount of processing power is required, and therefore, there is a possibility that the display is delayed only for the page in which the arrangement template 304 as shown in FIG. 43 is set.

  In order to provide the user with a page in which an image is pasted diagonally without performing such image conversion, as described above, in the square pasting area 441 including the display area 341, The image may be cut out, and the cut out image may be masked so that the image is displayed only in the display area 341. In this way, processing related to page display can be executed in the same processing as that for a page on which another layout template, for example, the layout template 301 (FIG. 41) is set. It is possible to perform processing without requiring a large processing capacity.

  In this way, by performing processing using the concept of two different regions, the display region and the pasting region, it is possible to reduce the processing capability related to image display and change the image to be provided to the user. (For example, a heart-shaped image as shown in FIG. 42 is provided).

  In this manner, the pasting area is a square regardless of the shape of the display area.

  The display area and the pasting area have a fixed size in the arrangement template. Since the pasting area has a predetermined size, if the image pasted on the pasting area is not the same size as the pasting area, some processing is required. This processing will be explained. In the above description, “processing” refers to processing described as “image clipping” or the like.

  As shown in FIG. 44, the still image stored in the still image DB 31 (FIG. 2) includes images having a plurality of aspect ratios. In the example shown in FIG. 44, the aspect of the image 511 is “16: 9”, the aspect of the image 512 is “3: 2”, and the aspect of the image 513 is “4: 3”. Although not shown in the figure, there are also images having other aspect ratios as the user processes the images. Thus, it is considered that there are various aspect ratios of still images recorded in the still image DB 31 (FIG. 2).

  The image includes a horizontally long image and a vertically long image. For example, as shown in FIG. 45, even if the image 521 and the image 522 both have an aspect ratio of “16: 9”, the image 521 is a horizontally long image, and the image 522 is a vertically long image. As described above, the still image recorded in the still image DB 31 (FIG. 2) includes a horizontally long image and a vertically long image (denoted appropriately as images having different directions).

  From such images having different aspect ratios and directions, an image having a size suitable for the pasting area is cut out.

  The image clipping will be described with reference to FIG. As shown in the upper diagram of FIG. 46, when an image having the maximum size in the form of the pasting area 411 is cut out from an image 511 having an aspect ratio of 16: 9, a part of the upper side of the image 511 and a lower side of the image 511 are displayed. An image with a part cut is cut out.

  “Cut out the image having the maximum size in the form of the pasting region 411 from the image 511” means that the image is cut out from the image 511 while maintaining the aspect ratio of the pasting region 411, and the image to be cut out is This means that clipping is performed so as to include the image 511 as much as possible.

  As will be described later with reference to FIG. 47, when the image is cut out, the size of the pasting area itself is not related, and the aspect ratio of the pasting area is concerned.

  46, when an image having the maximum size in the form of the pasting area 412 is cut out from an image 511 having an aspect ratio of 16: 9, a part on the left side and a part on the right side of the image 511 The image with the cut off is cut out.

  In this way, as shown in FIG. 46, even when an image suitable for the pasting area is cut out from the same image 511, if the aspect ratio of the pasting area is different, the area of the image 511 to be cut out is different.

  The aspect ratio of the pasting area will be further explained with specific numerical values.

  The upper diagram in FIG. 47 is a diagram showing the relationship between the arrangement template and the pasting area, and the lower diagram in FIG. 47 is an image pasted in the pasting area (hereinafter referred to as an original image as appropriate) and the pasting area. It is a figure which shows a relationship.

  47, the horizontal length of the arrangement template 541 is 1920 dots, and the vertical length is 1080 dots. On such an arrangement template 541 having an aspect ratio of “1920: 1080”, that is, an aspect ratio of “16: 9”, a pasting area having a horizontal length of 200 dots and a vertical length of 400 dots 542 is provided. The aspect ratio of the pasting area 542 is “200: 400”, that is, “1: 2”.

  When the image to be pasted on the pasting area 542 having the aspect ratio of “1: 2” is the original image 543 as shown in the lower diagram of FIG. 47, as shown in the lower diagram of FIG. An image is cut out from 543.

  The original image 543 has a horizontal length of 1600 dots and a vertical length of 900 dots. Therefore, the aspect ratio of the original image 543 is “16: 9”. Since the aspect ratio of the original image 543 is “16: 9” and the aspect ratio of the pasting area 542 is “2: 1”, it can be seen that the aspect ratio is different. When the aspect ratios are different as described above, control is performed so that an image having the maximum size is cut out from the original image 543 with the aspect ratio of the pasting area 542.

  As shown in the lower diagram of FIG. 47, when an area that maximizes the original image 543 is obtained while maintaining the aspect ratio “2: 1” of the pasting area 542, the horizontal length is 450 dots and the vertical length is 900. It can be seen that the dot area can be cut out. “450: 900” is “2: 1”.

  900 dots set as the vertical length is the same number of dots as the vertical length of the original image 543. That is, in this case, the maximum area (length) can be secured in the vertical direction of the original image 543.

  When the vertical direction is 900 dots, in order to maintain the aspect ratio “1: 2”, the horizontal direction needs to be 450 dots. Therefore, as described above, an image of “450 dots × 900 dots” having an aspect ratio of “1: 2” is cut out from the original image 543.

  In the example shown in FIG. 47, the size of the pasting area 542 is “200 dots × 400 dots”, and the size of the image to be cut out is “450 dots × 900 dots”. Therefore, when compared in size, the size of the pasting area 542 and the size of the image to be cut out are different sizes.

  As described above, in the present embodiment, an image having the same size as that of the pasting area 542 is not cut out from the original image 543, but an image having the same aspect ratio as that of the pasting area 542 is original. It is cut out from the image 543.

  Of course, in the case of the example shown in FIG. 47, since the cut-out image is larger than the pasting area 542, when the cut-out image is pasted on the pasting area 542, the image is reduced and pasted. Is done.

  By the way, when an image to be pasted from the original image is cut out based on the aspect ratio of the pasting area, only a part of the original image 543 may be cut out as described with reference to FIG. . In other words, not all of the original image is pasted in the pasting area, but only a part of the original image may be pasted.

  When only a part of the original image is cut out as described above, an inappropriate image may be provided to the user depending on the cutting position (region). For example, as shown in FIG. 48, when an image is cut out in the region of the pasting region 542 ′ maintaining the aspect ratio with respect to the original image 543 in which a human face is greatly projected, the cut out image 543 ′ is an image that includes only a part of the human face.

  The photographer who has photographed the original image 543 seems to have photographed the original image 543 with the intention of photographing around the face of a person. When the image 543 ′ cut out from the original image 543 shot with such intention includes only a part of a human face, or an image that does not include at all seems to have been cut out. In such a case, it is considered to be contrary to the intention of the photographer, and such an image can be said to be an inappropriate image.

  It is desirable that the image is cut out in a manner that reflects the intention of the photographer who captured the original image, not just the human face.

  Therefore, when clipping is performed, that is, when processing for pasting an image on an arrangement template is performed, the image that is clipped and pasted is made to reflect the intention of the photographer as much as possible. A description will be given of such processing.

[About image pasting]
The image pasting process executed in step S15 in the flowchart of FIG. 3 will be further described. As described above, the image displayed in the display area is cut out so that the photographer's intention is reflected. Here, a case where processing is performed while paying attention to the face of a person shown in an image will be described as an example.

  FIG. 49 shows an image in which a plurality of faces are projected on the original image 543. Of the plurality of faces, the face 601 is projected larger than the other faces (not labeled). In other words, even if an image is obtained by photographing a plurality of faces, the photographer intends to photograph the face 601 as long as only the face 601 is captured in the up position. It is thought that was taken. Therefore, in the case of such an original image 543, the pasting area 542 'is set so that the face 601 is included, and the image in the pasting area 542' is cut out from the original image 543.

  As an algorithm for cutting out an image, an algorithm that cuts out an area in the original image that includes the face that is projected most greatly is referred to as Algorithm 1. In other words, algorithm 1 is an algorithm that cuts out an area including a face of the largest area among faces of a predetermined size or larger in the original image.

  The algorithm 1 is an algorithm suitable for a case where one person is in a photograph or a small number of persons are in a photograph.

  FIG. 50 is a diagram for explaining the algorithm 2. In the original image 543 shown in FIG. 50, a plurality of faces (all of which are not labeled) are copied. The original image 543 shown in FIG. 50 and the original image 543 shown in FIG. 49 are images in which a plurality of faces are projected, but the original image 543 shown in FIG. 50 has a predetermined size or more. The difference is that the image has no face in the region.

  In this way, even when a plurality of faces are projected in the original image 543, in the case of an image in which there are no faces of a predetermined size or larger, control is performed so that clipping is performed in an area including as many faces as possible. Is done. Such an image in which a plurality of small faces are projected is an image such as a group photo. In such a group photo, a plurality of faces projected in the original image 543 are displayed. Of these, an image cut out in a region including as many faces as possible is considered to be an image that matches the photographer's intention.

  Algorithm 2 is an algorithm that cuts out an image from an original image in an area that includes as many faces as possible when there is no face of a predetermined size or larger in the original image. Algorithm 2 is an algorithm suitable for an image such as a group photo in which many people are shown.

  FIG. 51 is a diagram for explaining the algorithm 3. In the original image 543 shown in FIG. 51, a plurality of faces are shown. Of the projected faces, the faces 611, 612, and 613 are faces that are distorted, out of focus, or recognized as faces but are pictures. The face 614 is the face that is most clearly projected as compared to other faces, and can be determined as the face most likely to be the face.

  For example, there is an image in which the face 614 is in focus, but other portions including the face are not in focus. In addition, there is an image in which a photographer processes an image and distorts a part other than the face 614 or combines other images. When an image is cut out from the original image 543 in the region that was in the pasting region 542 ′, the region including the face that is most likely to be the face is set as the pasting region 542 ′, and the image in that region is the original. The image 543 is cut out.

  The algorithm 3 is an algorithm that cuts out an area including a face having a highest probability of being a face when it is determined that the original image 543 includes a plurality of faces.

  FIG. 52 is a diagram for explaining the algorithm 4. A plurality of faces are shown in the original image 543 shown in FIG. Of the projected faces, the face 621 is a face drawn in, for example, a painting. A face 622, a face 623, and a face 624 are each a live-action face.

  For example, when a photograph including a subject is photographed in front of a picture, the photograph is photographed as well as the subject. When a face is drawn in the painting, the face may be detected. The method for detecting a face from an image is described in the literature as described later, but it is considered that the detection accuracy differs depending on the method used. Depending on the detection accuracy of the technique used, it may be possible to accurately extract a drawn face such as a picture.

  As described above, when a face drawn in a picture or the like is extracted, an area including the face may be cut out. It can be said that a face drawn in a painting is the most unlikely face compared to a live-action face.

  When it is determined that the original image 543 includes a plurality of faces, the algorithm 4 sets an area including a face with the lowest probability of being a face as the pasting area 542 ′, and the pasting area 542 ′. This is an algorithm for cutting out the image from the original image 543.

  The algorithm 4 is an algorithm opposite to the algorithm 3. If clipping is performed only with algorithm 3, there is a possibility that an album in which the same person's face is continuously displayed may be created. It can be made an album in which the face of a painting is sandwiched between. Therefore, it is possible to provide the user with an album that is not a monotonous album (can provide a variety of albums).

  FIG. 53 is a diagram for explaining the algorithm 5. A plurality of faces are shown in the original image 543 shown in FIG. The algorithm 5 is an algorithm for setting a pasting area 542 ′ at random from the original image 543 and cutting it out.

  For example, algorithm 5 is applied to a case where even if a plurality of faces are detected, it cannot be processed by algorithms 1 to 4.

  Algorithms 1 to 4 are processes performed to reflect the photographer's intention as much as possible. However, when the algorithm 5 is applied, there is a possibility that an image according to the photographer's intention may not be provided. However, for example, when an image is cut out according to the algorithm 5, every time an album is browsed, the image is cut out based on the algorithm 5 and an album is created. Even so, because the way to cut out images is different (because it is cut out randomly), it seems that different albums have been created, and the album provided to the user has different contents for each viewing Is possible.

  Although there are people's preferences, some people think that albums always provide the same image, and some people want to provide such an album. Some people think that such an album is more interesting. Therefore, whether to apply the algorithm 5 may be set by the user viewing the album.

  Here, the algorithms 1 to 5 have been described as examples. However, these algorithms 1 to 5 are not descriptions showing limitations when an image is cut out. Of course, the cutout may be performed by an algorithm other than the algorithm 5 or 5. Furthermore, the algorithms 1 to 5 may be algorithms that are used alone or may be algorithms that are used in combination.

  Algorithms 1 to 5 are algorithms related to detecting a face and how to include the detected face in an image to be cut out. In order to apply such an algorithm, it is necessary to detect an area determined to be a face from an image. As a method for detecting a region determined to be a face from an image, for example, Japanese Patent Application No. 2004-93001 (Japanese Patent Laid-Open No. 2005-284348) and Japanese Patent Application No. 2004-94792 (Japanese Patent Application No. 2004-94792) previously filed by the present applicant. No. 2005-2844478) can be applied.

  Depending on the technique applied, not only can the face area be detected from the image, but it can be further detected whether the face is a male face or a female face (gender), or if it is a smiling face. It is possible to further detect whether the face is a facial expression (expression). In addition to face detection, when information associated with facial features such as gender and facial expressions (emotions that can be read from facial expressions) is also detected, such accompanying information is also used. Processing may be performed.

  For example, when information associated with a detected face is also acquired, an algorithm using the associated information may be referred to as algorithm 6, and an image may be cut out using the algorithm 6.

  When the face area is detected from the original image and information associated with the detected face is acquired, the algorithm 6 determines the condition of the specific face from the associated information. It is determined whether or not the face satisfies the condition, and when it is determined that the face satisfies a specific condition, the face is cut out from the original image so that the face is arranged in the center of the cut out image. It is an algorithm to do.

  The specific face condition is a condition determined from the accompanying information, and is, for example, the facial expression, the direction the face is facing, and sex as described above.

  For example, in the case where an image is cut out based on the algorithm 6, when a plurality of faces are detected from the original image, the pasting area is arranged so that the laughing face among the plurality of detected faces is arranged in the center. Is set and clipping is performed. Also, for example, when a male and female face is detected from the original image, clipping is performed so that the two are in the center.

  A process in which a plurality of algorithms are used to cut out an image suitable for the pasting area from the original image and paste the image to the pasting area of the arrangement template will be described with reference to the flowchart of FIG. The process described with reference to the flowchart of FIG. 54 is an image pasting process executed in step S15 of FIG.

  In step S111, face recognition processing is performed. This face recognition process is a process for detecting a face included in an original image to be processed by a predetermined method. As this predetermined method, the method described in the above-mentioned document can be used, and the description thereof is omitted here. Further, it goes without saying that the present invention can be applied even when face detection is performed by a method other than that described in the above-mentioned literature.

  In step S111, when an area where a face is reflected is detected from the original image to be processed, the process proceeds to step S112. In step S112, it is determined whether a face of a predetermined size or larger is detected among the detected face regions. This determination is a determination made to execute the algorithm 1 described above.

  If it is determined in step S112 that a face having a predetermined size or more is detected among the detected face regions, the process proceeds to step S113. In step S113, an area that includes a face of an area of a predetermined size or larger is set with the aspect ratio of the pasting area, and an image in the area is cut out. When a plurality of faces of a region of a predetermined size or more are detected, the region is set so that the face of the region of the maximum size is included, and the image is cut out.

  When the image is cut out, in step S114, the cut out image is pasted on the pasting area of the arrangement template that is the processing target.

  On the other hand, if it is determined in step S112 that no face of a predetermined size or larger is detected among the detected face areas, the process proceeds to step S115. In step S115, it is determined whether or not a predetermined number of faces have been detected. This determination is a determination made to execute the algorithm 2 described above.

  If it is determined in step S115 that the number of detected faces is greater than or equal to the predetermined number, the process proceeds to step S116. In step S116, an area that maximizes the number of faces is set according to the aspect ratio of the pasting area, and an image is cut out. When the image is cut out, in step S114, the cut out image is pasted on the pasting area of the arrangement template that is the processing target.

  On the other hand, if it is determined in step S115 that the number of detected faces is not greater than or equal to the predetermined number, the process proceeds to step S117. In step S117, it is determined whether a face corresponding to a specific condition has been detected. This determination is a case where the facial feature (accompanying information) can be acquired together with the detection of the face, and is a determination made to execute the algorithm 6. Alternatively, when the facial feature is not acquired, the determination is made to execute the algorithm 3 or the algorithm 4.

  When the determination in step S117 is performed in order to execute the processing based on the algorithm 3 or the algorithm 4, the specific condition is a condition that is determined to be a face most likely to be a face, or a face that is most unlikely to be a face. It is a condition determined to be.

  When the determination in step S117 is performed in order to execute the processing based on the algorithm 6, the specific condition is a face of laughing, a face facing the front, a man's face, a woman's face, etc. This is a condition that specifies the characteristics.

  When the image is cut out, in step S114, the cut out image is pasted on the pasting area of the arrangement template that is the processing target.

  On the other hand, if it is determined in step S117 that no face corresponding to the specific condition has been detected, the process proceeds to step S119. When the process has proceeded to step S119, a process related to clipping is executed based on the algorithm 5. That is, an image is cut out at random with the aspect ratio of the pasting area.

  When the processing comes to step S119, the face is included in the original image, but it is in a state where it is impossible to determine which face should be focused and the image should be cut out. For example, the control is performed so that a region including the central region of the original image is cut out.

  Even in the case where “the image is cut out randomly”, it is better to set a predetermined condition. For example, the image is cut out so as to include the central portion of the original image. Conditions such as cutting out so as to include an area are set, and cutting out is performed based on the conditions.

  When the image is cut out, in step S114, the cut out image is pasted on the pasting area of the arrangement template that is the processing target.

  Such processing is repeated by the number of images to be arranged in the arrangement template, thereby creating one page constituting the album. The created one page is, for example, the page 151 shown in FIG. 10 or the page 151 'shown in FIG.

  An album is created by repeating the creation for each page. The created album is, for example, the album 181 shown in FIG. The created album is recorded in the content DB 34 (FIG. 2).

[About viewing created albums]
Next, a description will be given of browsing of the album created in this way.

  FIG. 55 is a flowchart for explaining album display processing.

  In step S <b> 141, the recording / playback control unit 52 (FIG. 2) reads information related to the recorded album from the recording / playback unit 18. The recording / playback control unit 52 supplies display data for displaying a list of albums to the display control unit 53. Based on the control of the control unit 20, the image processing apparatus 1 displays a list of albums on the monitor 2.

  A display example of a list of albums displayed on the monitor 2 is shown in FIG. The album list display unit 701 is provided on the screen shown in FIG. The album list display section 701 displays an icon indicating each album, the album name on the right side of the icon, and the date and time when the image managed in the album is taken below the album name.

  In the album list display portion 701 of the screen shown in FIG. 56, six icons indicating albums 1 to 6, album names, and dates are displayed. In FIG. 56, the icons are shown as squares. However, for example, an image serving as an album cover may be used as an icon indicating each album.

  In FIG. 56, for example, it is displayed that the date and time when the album name “album 1” was photographed is “2005/3/15” to “2005/3/18”. Thus, the date and time displayed below each album name are the date and time when the oldest photo of the album was taken and the date and time when the newest photo was taken.

  On the monitor 2, a scroll bar 702 is displayed on the right side of the screen that is operated when displaying album icons that cannot be displayed in the album list display unit 701. Furthermore, a cursor 703 indicating which album is selected is also displayed. In the display example shown in FIG. 56, the cursor 703 is positioned on the album name icon “album 3”.

  Such a screen is displayed on the monitor 2 in step S141.

  In step S142, the selection unit 51 is selected by the user based on a signal indicating the user's operation for selecting one of the albums from the album list displayed on the monitor 2 supplied from the operation unit 19. Information indicating one of the albums is supplied to the recording / playback control unit 52 from the list of albums.

  For example, in the state of the screen shown in FIG. 56, that is, when the cursor 703 is on the “album 3” icon, when the user performs an operation for instructing the determination, the selected album is identified. Is generated and supplied to the record generation control unit 52.

  In step S <b> 143, the recording / playback control unit 52 supplies display data for displaying the selected album to the display control unit 53 based on information indicating one of the supplied albums. Based on the control of the control unit 20, the image processing apparatus 1 causes the monitor 2 to display the first page of the selected album.

  In step S144, the recording / playback control unit 52 determines whether there is a next page of the displayed album. If it is determined in step S144 that the next page does not exist, the process ends.

  On the other hand, if it is determined in step S144 that the next page exists, the process proceeds to step S145. In step S145, based on the control of the control unit 20, the image processing apparatus 1 causes the monitor 2 to display the next page of the selected album. By repeating the processes of step S144 and step S145, the pages constituting the album are sequentially provided to the user.

  Thus, the image processing apparatus 1 can display an album.

  Incidentally, FIG. 10 or FIG. 11 shows a display example displayed as one page of the album. Referring back to FIG. 11, three images are displayed on the page 151 ′. Among them, the image 171 and the image 172 are still images read from the still image DB 31 (FIG. 2). 161 has been described as a still image (thumbnail image) extracted from a moving image recorded in the moving image DB 32.

  Thus, in the present embodiment, a thumbnail image representing one scene of the moving image extracted from the moving image is also displayed as a still image on the page in the album. It has been described that, in order to perform such display, clustering is performed in consideration of the shooting date and time of moving images recorded in the moving image DB 32 when clustering. FIG. 35 shows an example of the clustered result. For example, an image captured at 0:13 on March 1 of group 1 may be a moving image.

  When clustering is performed, the moving image DB 32 is referred to and clustering is performed including the shooting date and time of moving images. As a result, the album provided to the user can include thumbnail images of moving images. Therefore, the description thereof is omitted here. Further, here, a description will be given of processing when the user associates a thumbnail image of a predetermined moving image with a predetermined album.

  The reason why the user can associate a thumbnail image of a desired moving image with a desired album is as follows.

  As described above, when clustering is performed in consideration of the shooting date and time of moving images, the thumbnail images of moving images can be displayed on a predetermined page as a result. Only moving images that are related to the image (in this case, related to being located in the vicinity in time) are displayed.

  For example, referring to FIG. 35 again, if the image taken at “01:13 on March 1” of group 1 is a thumbnail image extracted from the moving image, “March 1 in group 1” It is classified as Group 1 because it is a still image taken at 0:10 on the day, a still image taken at 0:12 on March 1, and a moving image taken close in time. It was. However, for example, when it is desired to associate a moving image taken on February 1 a month ago with group 1 or with an album including group 1, it is associated only with the clustering method described above. I can't.

  In order to respond to the user's desire to associate with the same group (the same album) even if the shooting date and time are separated, the user can select a thumbnail image of a desired moving image in a desired album. Can be associated. Such processing will be described with reference to the flowchart of FIG.

  In step S161, an instruction is received from the user to execute the moving image association process. FIG. 58 shows an example of a screen displayed on the monitor 2 when a moving image association is instructed. The screen shown in FIG. 58 is a screen displayed on the monitor 2 when the user instructs display of the menu screen in the screen shown in FIG. 56, for example.

  The menu screen refers to a screen in a state where the two menu items shown on the right side in FIG. 58 are displayed. That is, on the screen shown in FIG. 58, an item 731 “play album” and an item 732 “association of moving images” are displayed as menu items on the menu screen. In the case where the item 732 “association of moving images” is selected among these items, it is determined in step S161 that an instruction to execute the association processing of moving images is received from the user, and the process proceeds to step S162. Is advanced.

  58, when the item 131 “playback album” is selected, the album currently selected (the cursor 703 is positioned in the screen example shown in FIG. 58). (Album with the album name “album 3”) starts to be played. That is, the process based on the flowchart shown in FIG. 55 described above is started.

  In step S162, a moving image shot at a date close to the date of the selected album is searched. The search target is a moving image taken by the user, and does not include a moving image such as a television broadcast. In other words, only personal moving images are searched. Whether it is a personal moving image (whether it is a moving image having a copyright such as a television broadcast) can be determined by referring to metadata accompanying the moving image.

  Here, the description will be continued assuming that only a personal moving image is a search target, but a moving image such as a television broadcast can also be a search target.

  For example, in the screen example shown in FIG. 58, when the item 732 “association of moving images” is selected, “album selected” has an album name “album 3” where the cursor 703 is located. I have an album. Therefore, since the album having the album name “Album 3” has a date “2005/5 / 15-2005 / 5/18”, a moving image shot at a date close to this date is stored in the moving image DB 32. Searched.

  In step S163 (FIG. 57), the search result is used, and a list of searched moving images is displayed. As a result, a screen as shown in FIG. 59 is displayed on the monitor 2. A moving image list display section 751 is provided on the screen shown in FIG. The moving image list display section 751 includes an icon indicating each moving image, a moving image name on the right side of the icon, and the date and time when the moving image was shot below the moving image name (the date and time when shooting was started). And date and time of completion) are displayed.

  Note that in step S162, moving images shot at a date close to the date and time of the selected album are searched for when the searched moving images are displayed as a list in step S163. This is because the moving images are displayed in order of the shooting date and time of the images included in the images. This is because it is considered that there is a high possibility that a moving image close to the shooting date and time of an image included in the selected album is associated with the album.

  When displaying a list of moving images, for example, if the images are displayed in order of oldest shooting date or newest, the moving images are searched in step S162 so that the images are displayed in the set order.

  In the moving image list display portion 751 of the screen shown in FIG. 59, six icons, moving image names, and dates respectively indicating moving images 1 to 6 are displayed. In FIG. 59, the icon is shown by a rectangle, but for example, a thumbnail image indicating one scene of a moving image may be used as an icon indicating each moving image.

  In FIG. 59, for example, it is displayed that the date and time when the moving image name “moving image 3” was taken is “2005/5/16” to “2005/5/16”. As described above, the date and time displayed below each moving image name is the date and time when shooting of the moving image is started and the date and time when shooting is ended.

  On the monitor 2, a scroll bar 752 that is operated when displaying icons of moving images that cannot be displayed on the moving image list display section 751 is displayed on the right side of the screen. Further, a cursor 703 indicating which moving image is selected is also displayed. In the display example shown in FIG. 59, the cursor 703 is positioned on the icon having the moving image name “moving image 3”.

  Such a screen is displayed on the monitor 2 in step S163. The configuration of the screen on which the moving image list is displayed and the configuration of the screen on which the album list is displayed are basically the same as shown in FIGS. 59 and 58, respectively.

  When the screen as shown in FIG. 59 is displayed on the monitor 2 and the user performs a predetermined operation by operating the operation unit 19, the screen of the monitor 2 is displayed as a menu as shown in FIG. Switch to the screen. In the screen shown in FIG. 60, two menu items are displayed on the right side of the monitor 2. In the screen shown in FIG. 60, an item 771 “Determine” and an item 772 “Return” are displayed as menu items on the menu screen.

  When the item 771 “Determine” is selected among these items, it is determined in step S164 that the user has received an instruction to associate the selected moving image with the selected album. Processing corresponding to the instruction is executed.

  In FIG. 60, the cursor 703 is positioned on the icon indicating the moving image with the moving image name “moving image 3”. That is, the state of the screen shown in FIG. 60 is a state in which the moving image with the moving image name “moving image 3” is selected by the user as a target to be associated with the album. In order to make the user recognize the associated album, the album name “album 3” is displayed on the left side of the icon indicating “moving image 3”, and an album with the album name “album 3” is shown. An icon is displayed.

  In the screen shown in FIG. 60, when the item 771 “Determine” is operated, processing for associating “Album 3” with “Moving image 3” is executed in Step S164, and the item 772 “Return” is displayed. When operated, the screen of the monitor 2 is returned to FIG. 59 or FIG. 58, and the processing performed on those screens is executed.

  By performing such processing, the album and the moving image are associated with each other, and when the album is reproduced, the thumbnail image of the moving image is displayed in the reproduced album.

[About deletion of moving images]
By the way, a moving image has a larger data size than a still image. Therefore, when the recording capacity of the moving image DB 32 is not enough to record a new moving image, processing is performed to delete old moving image data and free up the capacity. Such processing may be performed according to a user instruction, or may be performed based on the determination of the image processing apparatus 1.

  As described above, the moving image recorded in the moving image DB 32 may be deleted. Of course, there is a possibility that the still image recorded in the still image DB 33 may be deleted, but the data size is smaller than that of the moving image. For example, there is a situation in which the still image must be deleted due to insufficient capacity. Therefore, it is considered that the probability of being deleted from the moving image is low. Therefore, the description will be continued here on the assumption that the moving image is deleted.

  As described above, a moving image is associated with an album at the time of clustering or when a user gives an instruction. If the associated moving image is deleted, a still image extracted from the moving image is stored. There is a problem that the still image is not displayed on the page on which the image is displayed.

  Here, the processing when a moving image is deleted will be described with reference to the flowchart of FIG. The process of the flowchart in FIG. 61 is executed when a deletion instruction is issued for a predetermined moving image among the moving images recorded in the moving image DB 32.

  In step S181, it is determined whether or not an associated album exists in the moving image to be deleted by the user. If it is determined in step S181 that there is no album associated with the moving image to be deleted, the process proceeds to step S182, and the moving image to be deleted is deleted from the moving image DB 32. Is done.

  On the other hand, if it is determined in step S181 that the associated album exists in the moving image to be deleted, the process proceeds to step S183. In step S183, the still image generated in advance from the moving image and recorded in association with the moving image is stored in association with the moving image in place of the moving image.

  As “a still image generated in advance from a moving image and associated with the moving image”, a thumbnail image indicating one scene of the moving image can be used. In addition, such a thumbnail image means that the thumbnail image is newly recorded in the still image DB 31 that “the album associated with the moving image is associated and stored instead of the moving image”. . In addition, when recorded in the still image DB 31, the shooting date / time of the moving image (date / time when shooting was started, date / time when shooting was ended, or date / time between the date / time when shooting was started and the date / time when shooting ended) is recorded. The date and time of the thumbnail image is recorded.

  In step S183, when a still image is recorded in the still image DB 31, the process proceeds to step S182, and the moving image to be deleted is deleted from the moving image DB 32.

  The processing related to moving image deletion described with reference to the flowchart of FIG. 61 will be briefly described again with reference to FIG. A moving image 161 ′ is recorded in the moving image DB 32, and a thumbnail image (still image 161) extracted from the moving image 161 ′ is recorded in association with the moving image 161 ′.

  In addition, a still image 171 and a still image 172 are recorded in the still image DB 31. In the state as shown in FIG. 62 (when the moving image 161 ′ is deleted), the page 151 ′ as shown in FIG. 11 is provided to the user as one page of the album. .

  When the moving image 161 ′ recorded in the moving image DB 32 is to be deleted, the still image 161 associated with the moving image 161 ′ is moved to the still image DB 31. Therefore, the still image DB 31 is in a state where the still image 171, the still image 172, and the still image 161 'are recorded.

  In this way, if a still image extracted from a moving image is recorded in the still image DB 31, the thumbnail image is also handled in the same manner as other still images and displayed on the album.

  Referring back to FIG. 11, FIG. 11 shows a display example of a predetermined page in the album. If there is a thumbnail image (still image 161) extracted from the moving image on the page 151 ′, “ It has already been described that the mark “Video” 162 is displayed.

  Even when the moving image 161 ′ that is the source of the still image 161 is deleted, the still image 161 itself is recorded in the still image DB 31 without being deleted by performing the above-described processing. It is displayed on the page 151 ′ together with the still image 171 and the still image 172. However, since the moving image 161 ′ has been deleted, a page 151 ′ in which the mark “Video” 162 is not displayed is provided to the user.

  Other processing related to moving image deletion will be described with reference to the flowchart of FIG. In the process described with reference to FIG. 61, there is a still image associated with a moving image in advance, and the still image is a process associated with an album instead of the moving image. The process described with reference is a process of generating a still image from a moving image and associating it with the album when the moving image is deleted.

  In step S201, it is determined whether or not an associated album exists in the moving image to be deleted by the user. If it is determined in step S201 that there is no album associated with the moving image to be deleted, the process proceeds to step S202, and the moving image to be deleted is deleted from the moving image DB 32. Is done.

  On the other hand, if it is determined in step S201 that the associated album exists in the moving image to be deleted, the process proceeds to step S203. In step S203, the scene change location of the moving image is detected. As a method for detecting a scene change from a moving image, for example, a method described in Japanese Patent Application No. 2002-254562 (Japanese Patent No. 3731657) filed earlier by the present applicant can be used.

  In step S203, when a scene change location of the moving image is detected from the moving image to be deleted, a still image corresponding to the detected scene change location is generated in step S204. In this case, since a still image is generated for each scene change location, as many still images as the number of scene change locations are generated.

  If there are many scene changes, the number of still images increases, so an upper limit is set for the number of still images to be generated, and the generated still images are thinned out according to the setting. Processing may be performed as appropriate.

  In this example, the scene change location of the moving image is detected and the still image is generated from the location. However, the still image may be generated from an arbitrary location designated by the user.

  Furthermore, when there are a plurality of still images generated, only the number of deleted moving images associated with the album (as a premise, a plurality of still images are extracted from one moving image, and the still images are extracted. May be extracted from the generated still image.

  In this way, the still image is generated from the deleted moving image because the data size of the still image is smaller than that of the moving image, so that the database may be recorded instead of the moving image. The user cannot include the thumbnail image of the moving image unless it is considered that the remaining amount of the moving image is not significantly affected, and the album associated with the moving image is not deleted even if the moving image is deleted. This is in order to respond to the desire to view the album in a new form.

  In step S204, the generated still image is recorded in the still image DB 31 in step S205.

  Thus, when a still image is generated from a moving image and recorded in the still image DB 31, the moving image to be deleted is deleted from the moving image DB 32 in step S202.

  Thus, when a still image is generated from a moving image and recorded in the still image DB 31, the still image is handled in the same manner as other still images recorded in the still image DB 31. Therefore, when the album is reproduced, the album including the still image generated from the moving image is reproduced.

  When a plurality of still images are generated and all the generated still images are included in the album, the generated plurality of still images are simply displayed at the place where the thumbnail image of one moving image was displayed. Since still images cannot be pasted, processing such as generation by re-editing an album including a plurality of still images is appropriately performed.

[About file management]
By the way, in the above-described embodiment, as shown in FIG. 2, the still image DB 31 and the moving image DB 32 are provided, and the still image and the moving image are managed in the respective databases. In addition, when still images and moving images are managed in the respective databases, if the moving images are deleted, the processing of the flowcharts shown in FIGS. 61 and 63 is executed.

  That is, since still images and moving images are managed separately in the respective databases, when a predetermined moving image is deleted from the moving image DB 32, a still image extracted from the deleted moving image is Processing such as newly recording in the image DB 31 was performed.

  Of course, the present invention can be applied not only when still images and moving images are managed in different databases, but also when still images and moving images are managed in the same database or in the same folder.

  When still images and moving images are managed in the same folder, for example, the data structure shown in FIG. 64 can be used. That is, in the folder configuration shown in FIG. 64, folder 1 and folder 2 are provided as folders. Folder 1 manages “still image — 01.jpg”, “still image — 02.jpg”, and “moving image — 01.mpg”, and folder 2 includes “still image — 03.jpg” and “still image — 04.jpg” , “Still image — 05.jpg”, “still image — 06.jpg”, and “moving image — 02.mpg” are managed.

  In this way, in the folder configuration example shown in FIG. 64, the folder 1 manages two still images and one moving image, and the folder 2 stores four still images and one moving image. I manage. As described above, still images and moving images may be managed in one folder. In other words, still images and moving images may be managed in the same folder or database, instead of managing still images and moving images in different folders and databases.

  In the folder configuration shown in FIG. 64, “moving image — 01.mpg” managed in folder 1 is to be deleted, and as described above, for example, the location of the scene change of “moving image — 01.mpg” When a still image is generated every time, after “moving image — 01.mpg” is deleted, the folder structure is changed as shown in FIG.

  When “moving image — 01.mpg” is deleted and “still image — 07.jpg” and “still image — 08.jpg” are generated from “moving image — 01.mpg”, as shown in FIG. Instead of “moving image — 01.mpg”, “still image — 07.jpg” and “still image — 08.jpg” are recorded in folder 1.

  As described above, when still images and moving images are managed in one folder, still images extracted from the deleted moving images are recorded instead of the deleted moving images. .

  Processing related to deletion of moving images when managing still images and moving images in one folder will be described with reference to the flowchart of FIG.

  In step S221, it is determined whether there is a folder associated with the moving image to be deleted by the user. If it is determined in step S221 that there is no folder associated with the moving image to be deleted, the process proceeds to step S222, and the moving image is deleted.

  On the other hand, if it is determined in step S221 that the associated folder exists in the moving image to be deleted, the process proceeds to step S223. In step S223, the scene change portion of the moving image is extracted. In step S224, a still image corresponding to the detected scene change location is generated. In step S225, the generated still image is recorded in the folder that managed the moving image instead of the moving image that is the deletion target.

  When a still image is generated from a moving image and recorded in the folder in this way, the moving image that is the object of deletion is deleted from the folder in step S222.

  As described above, in the present embodiment, when there is an album or folder associated with a moving image, when the moving image is deleted, a still image is generated from the moving image. Then, the generated still image is recorded instead of the moving image to be deleted.

  Even when the still image DB 31 and the moving image DB 32 as shown in FIG. 2 are provided and the still image and the moving image are managed in different databases, as described above (for example, As described with reference to the flowchart of FIG. 3, the album is created for images recorded in one folder. Therefore, even when moving images and still images are managed in a folder, an album can be created by the same processing as in the above-described embodiment.

  For example, in the file structure as shown in FIG. 64, when “Folder 1” is the target of album creation, the album including still images extracted from “moving image — 01.mpg” Created in the same way as However, in this case, since moving images and still images are managed in one folder, it is possible to omit processing (corresponding processing) of reading the shooting date and time of moving images with reference to the moving image DB 32. It is only necessary to read the shooting date and time of the image recorded in the folder.

  Further, even after “moving image — 01.mpg” is deleted, “still image — 07.mpg” and “still image” extracted from “moving image — 01.mpg” are stored in folder 1 as shown in FIG. Since the image_08.mpg ”is recorded, an album including these still images is created in the same manner as in the above-described embodiment.

  That is, according to the present invention, even when a moving image and a still image coexist in a folder, an album can be created by basically the same processing as described above.

  As described above, when data having a relatively large capacity (moving image data in the above example) is deleted, the data is converted to data having a relatively small capacity (still image data in the above example), and the capacity thereof. By using it instead of a large amount of data, as in the case of the album example described above, the data once created can be used effectively, or the data once presented to the user can be changed in content. Even so, the data can be presented again in a form close to the previously presented state.

  In addition, although an example of an album is shown here, the case of a moving image or a still image has been described as an example. As described above, when a relatively large amount of data is deleted, a relatively large amount of data is deleted. Converting to small data and using it instead of data having a large capacity can be applied to albums, moving images, still images, and the like without being limited to the above examples.

  In the above-described embodiment, the image processing apparatus 1 has been described as an example. For example, the file structure as shown in FIG. 64 is the same as the file structure when data is managed by, for example, a personal computer. That is, in a personal computer or the like, different types of data may be managed by one folder.

  Considering this, the processing described with reference to FIGS. 64 to 66, that is, when data with a relatively large capacity is deleted, it is converted into data with a relatively small capacity, and the data with the large capacity is stored. The process of being used instead of the process is not limited to the image processing apparatus 1 and can be applied to a personal computer or the like.

[About recording media]
FIG. 67 is a block diagram showing an example of the configuration of a personal computer that executes the series of processes described above by a program. A CPU (Central Processing Unit) 1001 executes various processes according to programs stored in a ROM (Read Only Memory) 1002 or a storage unit 1008. A RAM (Random Access Memory) 1003 appropriately stores programs executed by the CPU 1001 and data. The CPU 1001, ROM 1002, and RAM 1003 are connected to each other via a bus 1004.

  As the CPU 1001, the Cell described in “Birth of Cell”, Nikkei Electronics, Nikkei BP, Feb. 28, 2005, pages 89 to 117 can be used.

  An input / output interface 1005 is also connected to the CPU 1001 via a bus 1004. The input / output interface 1005 is connected to an input unit 1006 including a keyboard, mouse, microphone, and the like, and an output unit 1007 including a display, a speaker, and the like. The CPU 1001 executes various processes in response to instructions input from the input unit 1006. Then, the CPU 1001 outputs the processing result to the output unit 1007.

  A storage unit 1008 connected to the input / output interface 1005 includes, for example, a hard disk, and stores programs executed by the CPU 1001 and various data. A communication unit 1009 communicates with an external device via a network such as the Internet or a local area network.

  A program may be acquired via the communication unit 1009 and stored in the storage unit 1008.

  The drive 1010 connected to the input / output interface 1005 drives a removable medium 1011 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, and drives the program or data recorded therein. Get etc. The acquired program and data are transferred to and stored in the storage unit 1008 as necessary.

  The series of processes described above can be executed by hardware or can be executed by software. When a series of processing is executed by software, a program constituting the software executes various functions by installing a computer incorporated in dedicated hardware or various programs. For example, the program is installed in a general-purpose personal computer from the program storage medium.

  As shown in FIG. 67, a program storage medium for storing a program that is installed in a computer and can be executed by the computer is a magnetic disk (including a flexible disk), an optical disk (CD-ROM (Compact Disc-Read Only Memory), DVD (including Digital Versatile Disc), magneto-optical disk (including MD (Mini-Disc)), or removable media 1011 which is a package medium made of semiconductor memory, or the program is temporary or permanent ROM 1002 stored in the hard disk, a hard disk constituting the storage unit 1008, and the like. The program is stored in the program storage medium using a wired or wireless communication medium such as a local area network, the Internet, or digital satellite broadcasting via a communication unit 1009 that is an interface such as a router or a modem as necessary. Done.

  In the present specification, the step of describing the program stored in the program storage medium is not limited to the processing performed in chronological order according to the described order, but is not necessarily performed in chronological order. Or the process performed separately is also included.

  Further, in this specification, the system represents the entire apparatus constituted by a plurality of apparatuses.

  The embodiment of the present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the gist of the present invention.

It is a figure which shows the structure of one Embodiment of the image processing apparatus of this invention. It is a block diagram which shows the example of a structure of the control part of an image processing apparatus. It is a flowchart explaining the process which concerns on creation of an album. It is a figure which shows the example of a background template. It is a figure which shows the example of an arrangement | positioning template. It is a figure which shows the example of an arrangement | positioning template. It is a figure which shows the example of an arrangement | positioning template. It is a figure which shows the example of an arrangement | positioning template. It is a figure which shows the example of an arrangement | positioning template. It is a figure which shows the example of a structure of the predetermined page of an album. It is a figure which shows the other example of a structure of the predetermined page of an album. It is a figure which shows the example of a structure of an album. It is a figure which shows the example of a cluster. It is a figure which shows the example of a hierarchical structure. It is a figure explaining the browsing order of an image. It is a structural example of the part which performs the process regarding clustering. It is a figure explaining creation of a binary tree structure. It is a figure following FIG. 17 explaining preparation of a binary tree structure. It is a figure following FIG. 18 explaining preparation of a binary tree structure. It is a figure following FIG. 19 explaining preparation of a binary tree structure. It is a figure following FIG. 20 explaining preparation of a binary tree structure. It is a figure following FIG. 21 explaining preparation of a binary tree structure. It is a figure following FIG. 22 explaining preparation of a binary tree structure. It is a figure following FIG. 23 explaining preparation of a binary tree structure. It is a figure following FIG. 24 explaining preparation of a binary tree structure. It is a figure explaining the clustering based on grouping conditions. It is a figure which shows the example of the granularity of a cluster. It is a figure which shows the example of the setting of a group A division | segmentation flag. It is a figure which shows the example of the clustering result based on the grouping condition A. It is a figure which shows the example of the setting of a group B division | segmentation flag. It is a figure which shows the example of the clustering result based on the grouping conditions. It is a figure which shows the nesting relationship of a cluster. It is a flowchart explaining the process of clustering. It is a flowchart explaining the setting of a placement template. It is a figure explaining the processing result of clustering. It is a flowchart explaining the setting of a placement template. It is a figure explaining the relationship between imaging | photography space | interval and an arrangement | positioning template. It is a figure explaining the relationship between imaging | photography space | interval and an arrangement | positioning template. It is a flowchart explaining the setting of a placement template. It is a figure which shows the specific example of the relationship between imaging | photography space | interval and an arrangement | positioning template. It is a figure explaining the relationship between an aspect ratio and the sticking area | region of an image. It is a figure explaining the relationship between an aspect ratio and the sticking area | region of an image. It is a figure explaining the relationship between an aspect ratio and the sticking area | region of an image. It is a figure explaining the aspect-ratio of a still image. It is a figure explaining a vertically long and a horizontal still image. It is a figure explaining clipping of an image. It is a figure which shows the relationship between the aspect ratio of a sticking area | region, and the cutout area of an image. It is a figure explaining an improper cut-out area. It is a figure explaining the algorithm regarding clipping of an image. It is a figure explaining the algorithm regarding clipping of an image. It is a figure explaining the algorithm regarding clipping of an image. It is a figure explaining the algorithm regarding clipping of an image. It is a figure explaining the algorithm regarding clipping of an image. It is a flowchart explaining an image paste process. It is a flowchart explaining the display process of an album. It is a figure which shows the example of a screen displayed on a monitor. It is a flowchart explaining the process of correlation of a moving image. It is a figure which shows the example of a screen displayed on a monitor. It is a figure which shows the example of a screen displayed on a monitor. It is a figure which shows the example of a screen displayed on a monitor. It is a flowchart explaining the process regarding deletion of a moving image. It is a figure explaining the process regarding deletion of a moving image. It is a flowchart explaining the process regarding deletion of a moving image. It is a figure explaining the management of the image in a folder. It is a figure explaining the management of the image in a folder. It is a flowchart explaining the process regarding deletion of a moving image. It is a figure explaining a recording medium.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 Image processing apparatus, 2 Monitor, 18 Recording / reproducing part, 19 Operation part, 20 Control part, 31 Still image DB, 32 Moving image DB, 33 Template DB, 34 Content DB, 51 Selection part, 52 Recording / reproducing control part, 53 Display control unit, 54 content creation unit, 71 calculation unit, 72 template setting unit, 91 clustering unit, 92 hierarchy determination unit

Claims (7)

A management unit for managing templates prepared for each number of images to be arranged;
Acquisition means for acquiring information associated with the images to be grouped;
Grouping means for dividing the image into one or more groups based on the information obtained by the obtaining means;
Selecting means for selecting the template managed by the managing means according to the number of images included in the group generated by the processing of the grouping means;
An image processing apparatus comprising: arrangement means for arranging the image on the template selected by the selection means. The image processing apparatus according to claim 1, wherein the management unit manages a plurality of templates at different positions where the images are arranged for each number of the images. The image processing apparatus according to claim 1, wherein the grouping unit groups the number of images included in one group to be equal to or less than a maximum number of images that can be arranged by one template managed by the management unit. . The image processing apparatus according to claim 1, wherein the information is information related to a date and time when the image is captured. The selection unit refers to the shooting time of the images included in the group, calculates a shooting interval between images, compares the shooting time with a predetermined threshold, and selects the template according to the comparison result. The image processing apparatus according to claim 1. A management step for managing templates prepared for each number of images to be arranged, which is a template for determining the arrangement of images;
An acquisition step of acquiring information associated with the images to be grouped;
A grouping step of dividing the image into one or more groups based on the information acquired in the processing of the acquisition step;
A selection step of selecting the template managed in the management step according to the number of images included in the group generated in the grouping step;
An image processing method comprising: an arrangement step of arranging the image on the template selected in the processing of the selection step. A management step for managing templates prepared for each number of images to be arranged, which is a template for determining the arrangement of images;
An acquisition step of acquiring information associated with the images to be grouped;
A grouping step of dividing the image into one or more groups based on the information acquired in the processing of the acquisition step;
A selection step of selecting the template managed in the management step according to the number of images included in the group generated in the grouping step;
A program that causes a computer to execute a process including an arrangement step of arranging the image on the template selected in the process of the selection step.

Images