JP2016122413A - Image processing apparatus, control method of image processing apparatus, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To select an image in common with each of image sets.SOLUTION: There is provided an image processing apparatus comprising: a determination part that determines an attribute of a key image from an attribute of a subject present in common in a plurality of image sets; a key image selection part that selects a key image from each of the image sets by using the determined attribute; and a non-key image selection part that selects non-key images that are images other than the key image from each of the image sets.SELECTED DRAWING: Figure 6

Description

  The present invention relates to an image processing apparatus that selects a representative image common to each image set from a plurality of image sets, a control method for the image processing apparatus, and a program.

  In recent years, digital cameras have become widespread, the chances of taking images and moving images have increased due to the increase in storage media capacity and price, and the storage amount of images has also increased. For example, images of events such as Christmas, birthdays, and athletic events are selected from images stored in large quantities, and albums and movies are created.

  Patent Document 1 discloses a technique that facilitates selection of an image by using a template according to an event.

  Patent Document 2 discloses a technique for selecting a representative image showing a subject symbolizing an event, that is, a key image from a set of images taken at one event.

JP 2005-184790 A JP 2011-215963 A

  However, as in Patent Document 1, when the condition for selecting as a key image is fixed in advance, if the event is repeated multiple times such as Christmas or birthday, the same type of key image is common to each image set. Does not necessarily exist. When creating a single album by collecting a plurality of sets of images of the same type of event, it is not possible to select a key image common to each event, and it is not possible to feel the commonality among the events.

  Also, as in Patent Document 2, when selecting a representative image of each image set according to the score of the image, different types of subjects may be selected in each image set, and when they are collected to create one album , I can not feel the commonality between the representative images.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide a technique for selecting an image having commonality among image sets.

As a means for achieving the above object, an image processing apparatus according to the present invention comprises the following arrangement. That is,
Determining means for determining the attribute of the key image from the attributes of the subject that exist in common through a plurality of image sets;
Key image selection means for selecting the key image from each image set using the determined attribute;
Non-key image selection means for selecting a non-key image that is an image other than the key image from each of the image sets.

  According to the present invention, it is possible to select an image that is common to each image set.

1 is a diagram illustrating an example of a hardware configuration of an image processing apparatus according to an embodiment of the present invention. The example of a screen of the album creation application which concerns on one Embodiment of this invention. The block diagram of the album creation application which concerns on one Embodiment of this invention. The image management table which concerns on one Embodiment of this invention. The event table which concerns on one Embodiment of this invention. 6 is a flowchart showing a procedure of album creation processing performed by the image processing apparatus according to the embodiment of the present invention. The album creation condition table which concerns on one Embodiment of this invention. The album creation information table which concerns on one Embodiment of this invention. 6 is a flowchart illustrating a processing procedure for determining a key tag (key attribute) performed by the image processing apparatus according to the first embodiment. The tag detection table which concerns on 1st Embodiment. 6 is a flowchart showing a common tag detection processing procedure performed by the image processing apparatus according to the first embodiment. The comparison table which concerns on 1st Embodiment. The figure which shows the example of the creation album which concerns on 1st Embodiment. The figure which shows the example of the creation movie which concerns on 1st Embodiment. 9 is a flowchart illustrating a processing procedure for determining a key tag (key attribute) performed by the image processing apparatus according to the second embodiment. The tag detection table which concerns on 2nd Embodiment. 10 is a flowchart illustrating a processing procedure for determining a key tag (key attribute) performed by the image processing apparatus according to the third embodiment. 10 is a flowchart illustrating a processing procedure for determining a key tag (key attribute) performed by the image processing apparatus according to the fourth embodiment.

  Hereinafter, the present invention will be described in detail based on preferred embodiments with reference to the accompanying drawings. The configurations shown in the following embodiments are merely examples, and the present invention is not limited to the illustrated configurations.

<Embodiment 1>
In this embodiment, an image processing apparatus that selects an event type and automatically creates an album will be described as an example. As an example of the image processing apparatus according to the present invention, an album creating application that operates on a personal computer (PC) will be described as an example.

  FIG. 1 is a block diagram illustrating an example of a hardware configuration of the image processing apparatus 100 according to the present embodiment. The CPU 101 is involved in all the processes of each configuration, reads the instructions stored in the ROM 102 and RAM 103 in order, interprets them, executes the processes according to the results, and performs various controls.

  The recording device 104 stores a large amount of data such as an HDD or an SSD, and is used to store a material for creating an album, a creation result, and a program for executing the image processing. An input / output interface (I / F) 105 is an interface necessary for connection with an input / output device, and is used for connection with a display 106, a keyboard 107, a mouse 108, and a network interface (I / F) 109. .

  The display 106 is an output device for displaying an operation screen. A keyboard 107 and a mouse 108 are input devices for receiving an operation input by a user. The network interface 109 is used for connecting to an external device through a network.

  Note that the hardware configuration of the image processing apparatus 100 is not limited to the above-described configuration. For example, the display 106 may include a touch sensor and detect an input via the touch sensor. Further, it may be a device connected to a network such as a cloud service. In this embodiment, the storage location of the material is the recording device 104, but a storage device may be provided outside the image processing device 100 via the network interface 109.

  FIG. 2A illustrates an example of an operation screen of the album creation application. Reference numeral 201 denotes an operation screen of the album creation application. The operation screen 201 includes an event selection button 202, a creation button 203, a preview area 204, an output button 205, an end button 206, and the like.

  When the event selection button 202 is pressed, an event type indicating an event type included in an image management table described later and a list of image sets having the event type are displayed as in the example shown in FIG. Is done. The user selects an event type to be created in the album and an image set having the event type. The image set is classified for each shooting time and is a set of images having the same event type. A classification method will be described later. When an event type and an image set are selected, an image set belonging to the selected event type is selected from an image management table described later.

  The creation button 203 is a button for starting creation of an album with the event type selected by the event selection button 202 and an image set having the event type. The preview area 204 is an area for displaying the created album. When the output button 205 is pressed, the album creation information table describing the album conditions is output to an arbitrary location. Details of the album creation information table will be described later. An end button 206 is a button for ending the album creation application.

  Next, FIG. 3 is a diagram illustrating a functional configuration of the image processing apparatus 100 according to the present embodiment, that is, a configuration of an album creation application. The image database 300 stores an image management table for managing a list of images used for album creation.

  Here, FIG. 4 shows an example of the image management table. The image managed by the image management table includes an image ID 401 for identifying an image, a path name 402 indicating a file storage destination, a shooting date and time 403, an event type 404, a tag 405 indicating the type and attribute of the subject, and a score obtained by scoring image quality. 406, associated with a group ID 407 that identifies the image set. As for the event type 404, a registered event type of a kind is given in advance when storing an image. A known method may be applied as a method for assigning an event type. For example, an event is determined by a method disclosed in Japanese Patent Application Laid-Open No. 2010-81453, and may be automatically assigned, or may be assigned by a user. .

  Similarly, with respect to the tag 405, a pre-registered type tag is assigned at the time of image storage. As a tag addition method, it may be automatically applied using existing face recognition, object recognition, subject extraction, or the like, or may be given by a user. Also, a known method may be applied to the score calculation processing of the score 406, and for example, the score may be calculated by a method disclosed in Japanese Patent Application Laid-Open No. 2008-276668. In this embodiment, the score is scored between 0 and 1. The group ID 407 is clustered based on the event type and the shooting date and time when the image is stored, and may be automatically assigned to the classified image set, or may be manually added by the user.

  Subsequently, the event type acquisition unit 301 acquires the event type information selected by the user from the list of event types displayed by pressing the event selection button 202. The image set extraction unit 302 extracts an image set corresponding to the event type acquired by the event type acquisition unit 301. The event database 303 stores an event table to which a tag list that is a tag candidate indicating a subject symbolizing an event is assigned for each event type.

  FIG. 5 shows an example of the event table. The event type 501 is an identifier indicating the type of event such as birthday, Christmas, or travel, and a tag 502 described in the event table represents a subject symbolizing the event, and a plurality of candidates for each event type has a priority. It is assigned as a tag list with. In addition, the tag in the tag list described in the event table, the number of tags, and the priority are not fixed, and they may be changeable.

  The key tag determination unit 304 determines a key tag (key attribute) that is a tag (attribute) of a subject to be included in the key image from the event type and the extracted image set. A key image is a representative image in which a subject symbolizing an event is shown.

  The key image selection unit 305 selects an image having the key tag determined by the key tag determination unit 304 from each image set. The non-key image selection unit 306 selects an image having no key tag from each image set. The selected image output unit 307 arranges the images output by the key image selection unit 305 and the non-key image selection unit 306 based on the creation condition table stored in the creation condition DB 308, and outputs it as an album creation information table To do.

  The creation condition table is stored in the creation condition DB 308 and prepared for each spread page. As shown in FIG. 7, in the creation condition table, as image conditions 701, key images and non-key images (non-key images 1 to 4 in the example of FIG. 7), the number of non-key images (four in the example of FIG. 7). ) And the coordinates of the display area in the page for each image are described as coordinates 702. In the present embodiment, it is assumed that one image set is assigned to one spread page. Hereinafter, the description regarding the page will be described as a spread page.

  Next, with reference to the flowchart of FIG. 6, the procedure of the album creation process performed by the image processing apparatus according to the embodiment of the present invention will be described. In step S <b> 601, the event type acquisition unit 301 acquires event type information selected by the user from a list of event types displayed by pressing the event selection button 202.

  In step S602, the image set extraction unit 302 extracts an image set corresponding to the event type from the image management table (FIG. 4) using the event type information acquired in step S601. At this time, the total number of extracted image sets is acquired as the total number n of image sets. The extracted image set is arranged as S0, S1, S2,..., Sn-1 in order of photographing date and time, that is, in order of time when the event is performed.

  In step S603, the key tag determination unit 304 determines a key tag K that is common to the image sets from the image set group extracted in step S602 and the acquired event type. Details will be described later with reference to the flowchart of FIG. In S604, the image set extraction unit 302 acquires the first image set S0 extracted in S602, and sets the image set S0 as the current image set S.

  In step S <b> 605, the key image selection unit 305 selects an image with a high score from the images including the key tag K determined in step S <b> 603 from the current image set S as a key image. Here, when there is no image including the key tag K, an image having the highest score in the current image set S may be an image including the key tag K. Alternatively, an image selected by applying the priority of the tag in the tag list of the event table (FIG. 5) corresponding to the selected event type may be an image including the key tag K. Further, an image selected by a combination thereof may be an image including the key tag K.

  In step S <b> 606, the non-key image selection unit 306 selects an image that does not include the key tag K from the current image set S. The number of images to be selected here is the number of non-key images obtained from the creation condition table of FIG. As a method for selecting an image other than the key image, the images may be selected at random, in the order of image quality score, or may be selected so that the tags do not match. Moreover, you may apply and apply the priority of the tag of the tag list | wrist of the event table (FIG. 5) corresponding to the selected event type. Or you may combine each selection method.

  In step S <b> 607, the image set extraction unit 302 determines whether the processing has been completed for all the extracted image sets, that is, whether the current image set is Sn−1. When it is determined that the processing has been completed for all image sets (S607; Yes), the process proceeds to S608. On the other hand, if it is determined that the processing has not been completed for all image sets (S607; No), the process proceeds to S609.

  In S608, the selected image output unit 307 arranges the selected image on each page in the album, and outputs album creation information. An image is arranged for each page based on the coordinates described in the creation condition table (FIG. 7). The coordinates specify the upper left and lower right of the display area starting from the upper left of the page. In the creation condition table (FIG. 7), a plurality of types are stored in a state where IDs are assigned to the creation condition DB 308. The album creation information is recorded in the album creation information table shown as an example in FIG. The album creation information table manages the ID of the creation condition table and the ID of the selected image in association with each other. An album is created based on the album creation information table.

  In step S609, the image set extraction unit 302 changes the image set S to be processed to the next image set. Thereafter, the process returns to S605. Thus, each process of the flowchart of FIG. 6 ends.

  Next, with reference to a flowchart of FIG. 9, a procedure of processing for determining a key tag (processing of S603) performed by the image processing apparatus 100 according to the present embodiment will be described. In step S901, the key tag determination unit 304 acquires a list of tags corresponding to the event type acquired in step S601 from the event table (FIG. 5).

  In step S902, the key tag determination unit 304 acquires key tag candidates from the tag list acquired in step S901 in descending order of priority, and creates a tag detection table. FIG. 10 shows an example of a tag detection table. If the event type acquired in S601 is “Christmas”, the tag list corresponding to the event type in FIG. 5 is acquired. Key tag candidates are arranged from the tag list in descending order of priority from the top.

  In the example of FIG. 10, the key tag candidates are arranged in the order of “Christmas tree” (first priority), “present” (second priority), “cooking” (third priority), and “cake” (fourth priority). . In the image set S0, S1, S2,..., Sn-1, a case where an image with the tag is found is “◯”, a case where the image is not found is “×”, and in S902, all are “×”. To "".

  In step S903, the key tag determination unit 304 acquires the first image set S0 and sets the image set S0 as the current image set S. In step S904, the key tag determination unit 304 sets all images in the current image set S to I0, I1, I2,... Im-1 (m = total number of images) in the ID order, and sets the first image I0 as Set as current image I.

  In S905, the key tag determination unit 304 can represent each tag as T0, T1, T2,..., Tl−1, assuming that the number of all tags assigned to the current image I is l. T is set as the first tag T0.

  In step S906, the key tag determination unit 304 determines whether or not the current tag T is included in the tag list acquired in step S901. If it is included (S906; Yes), the process proceeds to S907. On the other hand, if it is not included (S906; No), the process proceeds to S908.

  In S907, the key tag determination unit 304 records that the current tag T is detected in the current image set S in the tag detection table (FIG. 10) created in S902. In FIG. 10, when the key tag candidate vertically aligned with the current tag T matches, the portion where the key tag candidate and the current image set S intersect is recorded as “◯”.

  In step S908, the key tag determination unit 304 determines whether all tags assigned to the current image I have been detected. If all tags have been detected (S908; Yes), the process proceeds to S910. On the other hand, when not all tags have been detected (S908; o), the process proceeds to S909.

  In step S909, the key tag determination unit 304 changes the target tag T assigned to the current image I to the next tag. In step S <b> 910, the key tag determination unit 304 determines whether the processing has been completed for all the images included in the current image set S. When the process is complete (S910; Yes), the process proceeds to S912. On the other hand, if the process has not ended (S910; No), the process proceeds to S911.

  In step S911, the key tag determination unit 304 changes the image I to be detected by the tag to the next image in the image set S. In S912, if the key tag determination unit 304 has finished the process of determining whether or not the process has been completed for all the extracted image sets (S912; Yes), the process proceeds to S914. On the other hand, if the process has not ended (S912; No), the process proceeds to S913.

  In S913, the key tag determination unit 304 changes the image set S to be detected to the next image set. In step S <b> 914, the key tag determination unit 304 analyzes the tag detection table (FIG. 10) and detects a tag common to each image set. Details will be described later with reference to the flowchart of FIG. In step S <b> 915, the key tag determination unit 304 determines a common tag as the key tag K. Thus, each process of the flowchart of FIG. 9 ends.

  Next, with reference to the flowchart of FIG. 11, the details of the process of detecting a common tag (S914) performed by the image processing apparatus 100 according to the present embodiment will be described.

  In step S1101, the key tag determination unit 304 creates a comparison table (FIG. 12) for comparing the number of detected tags, and initializes the value of the detected number of each tag to zero. As shown in FIG. 12, the comparison table includes tags 1201 described in the tag detection table (FIG. 10) and a total number (detection number) 1202 in which tags are detected for each image set.

  In S1102, the key tag determination unit 304 acquires the tag T having the highest priority from the tag detection table (FIG. 10), and sets the current tag T as the first tag T0. In step S <b> 1103, the key tag determination unit 304 acquires the first image set S in the tag acquired in step S <b> 1102 and sets it as the current image set S.

  In S1104, the key tag determination unit 304 confirms whether or not the tag T exists in the tag detection table (FIG. 10) for the current image set S. If there is, the key tag determination unit 304 determines the corresponding tag in the comparison table (FIG. 12). Add 1 to the field (number of detections).

  In step S1105, the key tag determination unit 304 determines whether or not the detection processing for all image sets extracted in step S602 has been completed. When the detection process is completed (S1105; Yes), the process proceeds to S1107. On the other hand, if the detection process has not ended (S1105; No), the process proceeds to S1106.

  In S1106, the key tag determination unit 304 changes the target image set S to the next image set. In step S1107, the key tag determination unit 304 determines whether or not processing has been completed for all tags in the tag detection table (FIG. 10). If processing has not been completed for all tags (S1107; No), the process proceeds to S1108. On the other hand, if the processing has been completed for all the tags (S1107; Yes), the process proceeds to S1109.

  In step S1108, the key tag determination unit 304 changes the tag T to a tag having the next highest priority. In step S1109, the key tag determination unit 304 determines whether there are tags detected for all (n) image sets in the comparison table (FIG. 12), that is, tags that exist in common in all image sets. To do. If it is determined that there is a common tag in the comparison table (S1109; Yes), the process proceeds to S1110. On the other hand, if it is determined that there is no common tag in the comparison table (S1109; No), the process proceeds to S1111.

  In S1110, the key tag determination unit 304 determines, as a common tag (key tag K), a tag having the highest priority among tags detected in common in all image sets using the comparison table (FIG. 12).

  In S1111, the key tag determination unit 304 uses the comparison table (FIG. 12) to determine, as a common tag (key tag K), a tag with a large number of image sets detected for each tag and a high priority. Thus, each process of the flowchart of FIG. 11 is completed.

  Here, an example of the created album is shown in FIG. A page 1301, a page 1302, and a page 1303 represent Christmas pages of 2011, 2012, and 2013, respectively. An image 1304, an image 1305, and an image 1306 each represent a key image for each year. An image 1307 to an image 1309, an image 1310 to an image 1314, and an image 1315 to an image 1318 respectively represent non-key images other than the key image for each year. As shown in FIG. 13, in order to emphasize the key image, the key image display area is arranged larger than the non-key image display area for each page to which each image set is assigned. May be.

  Further, regarding the arrangement position of the display image of the key image, it may be arranged for each page in the area that becomes the head of the page. Regarding the arrangement of non-key images, the same ID creation conditions table (FIG. 7) may be used to arrange images of the same type for each page at the same position so that the pages have uniformity. Further, as shown in FIG. 13, diversity may be provided between pages by using a creation condition table having different IDs for each page.

  In this embodiment, an example of automatically creating an album has been described, but a movie may be automatically created. The number of pages corresponds to the number of seconds of the time length of the movie to be generated. FIG. 14 shows an example of the generated movie. The movie includes a scene 1401, a scene 1402, and a scene 1403 representing events of each year. For each scene, a key image 1404, a key image 1405, and a key image 1406 are arranged, and a non-key image 1415 is arranged from the non-key image 1407.

  Further, increasing the display area of the key image in the album corresponds to extending the display time in the movie. Similarly, placing the key image at the top of the page in the album corresponds to placing the key image at the top of the scene divided for each event in the movie. Thus, the spatial arrangement of the images on the album page corresponds to the arrangement on the time axis in the movie. In this embodiment, an example in which a still image is used as a material used for an album or a movie has been described. However, if there is a tag indicating a subject, a moving image may be used.

  In this embodiment, one image set is assigned to one spread page. However, the present invention is not limited to this, and one image set may be assigned to a plurality of pages. Moreover, although the number of key images allocated per page is an example of one image, there may be a plurality of types or a plurality of types.

  As described above, according to the present embodiment, for an event that repeats a plurality of times such as Christmas or birthday, when collecting an image set of those arbitrary events and creating one album, An image that symbolizes the event and shows a subject common to each image set can be selected as a key image from the image set every year.

  That is, it is possible to select a key image showing a common subject symbolizing an event through all image sets. A common key image can make the entire album uniform, and an album with a story can be created. In addition, by providing a plurality of candidates in the event table, other key image candidates can be selected as key images even when there is no key image common to each image set.

<Embodiment 2>
In the first embodiment, the example in which the common tag used as the key image is determined using the presence / absence of tag detection and the priority prepared in advance from the common tags is described. On the other hand, in this embodiment, an example in which a common tag is determined using the total number of detected tags will be described.

  In the second embodiment, the common tag detection flowchart of FIG. 11 in the first embodiment replaces the common tag detection flowchart of FIG. Since the configuration of the image processing apparatus 100 is the same as that of the first embodiment, the description thereof is omitted. Unlike the case where the tag detection table (FIG. 10) in FIG. 10 is recorded in the tag detection table in the first embodiment, in the tag detection table (FIG. 16) in the present embodiment, in S907 in FIG. The number of key tag candidate tag detections is recorded for each image set.

  Hereinafter, with reference to the flowchart of FIG. 15, the details of the process of detecting a common tag (S914) performed by the image processing apparatus 100 according to the present embodiment will be described. In step S1501, the key tag determination unit 304 creates a comparison table for comparing the total number of detected tags and a variable Max for comparing the maximum value of the total number, and sets the value of the total number and the value of the variable Max. Initialize to 0. The comparison table according to the present embodiment includes the tags described in the tag detection table (FIG. 16) and the total number in which each tag is detected.

  In S1502, the key tag determination unit 304 acquires a tag T having a high priority from the tag detection table (FIG. 16) as in S1102, and sets the current tag T as the first tag T0. In S1503, the key tag determination unit 304 acquires the first image set S0 in the tag acquired in S1502, and sets the image set S0 as the current image set S.

  In step S <b> 1504, the key tag determination unit 304 acquires the number of detected tags from the tag detection table, and determines whether the number of detections is 1 or more. When one or more tags are detected (S1504; Yes), the process proceeds to S1505. On the other hand, if the number of detected tags is 0 (S1504; No), the process proceeds to S1506.

  In step S1505, the key tag determination unit 304 adds the number of detected tags to the item of the current tag T as a target in the comparison table created in step S1501. In step S1506, the key tag determination unit 304 sets the value of the item of the current tag T to be a target in the comparison table created in step S1501 to 0.

  In step S <b> 1507, the key tag determination unit 304 determines whether the detection processing for all the extracted image sets has been completed. When it is determined that the detection process has been completed (S1507; Yes), the process proceeds to S1509. On the other hand, if the detection process has not ended (S1507; No), the process proceeds to S1508.

  In step S1508, the key tag determination unit 304 changes the target image set S to the next image set. In step S1509, the key tag determination unit 304 determines whether the processing has been completed for all tags included in the tag detection table. When the processing is completed for all tags (S1509; Yes), the process proceeds to S1511. On the other hand, if the processing has not been completed for all tags (S1509; No), the process proceeds to S1510.

  In step S1510, the key tag determination unit 304 changes the target image set S to the next image set. In step S1511, the key tag determination unit 304 substitutes the detected total number of the first tag T from the comparison table into the variable Max. In step S1512, the key tag determination unit 304 changes the tag T to the next tag. In step S1513, the key tag determination unit 304 compares the number of detected tags T with the number of detected variables Max, and substitutes the larger number for the variable Max.

  In step S1514, the key tag determination unit 304 determines whether the processing has been completed for all tags. When it is determined that the processing has been completed for all the tags (S1514; Yes), the process proceeds to S1515. On the other hand, if it is determined that the processing has not been completed for all tags (S1514; No), the process returns to S1512.

  In step S1515, the key tag determination unit 304 determines whether or not the value of the variable Max is 0. If the value of Max is not 0 (S1515; Yes), the process proceeds to S1516. On the other hand, if the value of Max is 0 (S1515; No), the process proceeds to S1517.

  In step S1516, the key tag determination unit 304 determines a tag having a value of Max as a common tag. When there are a plurality of tags having a value of Max, the tags are determined in descending order of the priority of the tags in the event table.

  In S1517, the key tag determination unit 304 compares the number of image sets detected for each tag as in S1111 and determines a tag with a large number of detected image sets and a high priority as a common tag. Thus, the processes in the flowchart of FIG. 15 are completed.

  As described above, in the present embodiment, the key tag is determined by comparing the number of tags detected in common. As a result, it is possible to select, as a key image, an image of a subject that is highly interested in the user because of the large number of subjects photographed by the user.

  In the case where there is no common tag (S1515; No), the example in which the common tag is determined by the same method as in the first embodiment has been described. The common tag may be determined using the number of detected tags.

<Embodiment 3>
In the second embodiment, the example in which the common tag used as the key image is determined based on the number of common tags detected has been described. In contrast, in the present embodiment, an example in which a common tag is determined based on the image quality score will be described. In the third embodiment, the common tag detection flowchart of FIG. 15 in the second embodiment is replaced with the common tag detection flowchart of FIG. Since the configuration of the image processing apparatus 100 is the same as that of the first embodiment, the description thereof is omitted.

  Also, the description of the same part as in FIG. Note that the comparison table used in the present embodiment includes the tags described in the tag detection table and the image quality score average value calculated for each tag.

  Hereinafter, with reference to the flowchart of FIG. 17, the details of the process of detecting a common tag (S914) performed by the image processing apparatus 100 according to the present embodiment will be described. In step S1701, the key tag determination unit 304 extracts an image having the tag T in the current image set S from the tag detection table, and acquires the maximum image quality score from the image quality management table. The image quality management table manages tags and image quality scores of a plurality of images related to the tags in each image set.

  In S1702, the key tag determination unit 304 adds and records the maximum score value acquired in S1701 to the comparison table. In step S1703, the key tag determination unit 304 acquires the value of the image quality score added for each image set from the comparison table, divides by the total number of image sets, calculates the score average value, and records the calculated value in the comparison table. To do.

  In step S1704, the key tag determination unit 304 sets the score value of the corresponding image set S in the comparison table to 0.

  In step S1705, the key tag determination unit 304 substitutes the score average value of the first tag of the corresponding tag T from the comparison table into the variable Max. In step S <b> 1706, the key tag determination unit 304 compares the score average value of the tag T with the score average value of the variable Max, and substitutes the larger value into the variable Max. Above, description of each process of the flowchart of FIG. 17 is complete | finished.

  As described above, in the present embodiment, the maximum value of the image quality score is acquired from each image set among the tags that are present in common, and the average value of the acquired maximum values is calculated. A tag having a high average value among the common tag candidates is set as a key tag. Thereby, a key image can be selected with good image quality regardless of the number of detections.

  When there is no common tag (S1515; No), the common tag is determined by the same method as in the first embodiment, but not only the number of presence / absence of tag detection but also added to the image. The common tag may be determined using the size of the score obtained.

<Embodiment 4>
In the third embodiment, an example in which a common tag is determined by calculating an average value for each tag with respect to the maximum image quality score of each image set has been described. On the other hand, in this embodiment, an example will be described in which a common tag is determined using a minimum value instead of an average value. In the fourth embodiment, since the configuration of the image processing apparatus 100 in which the flowchart of FIG. 17 in the third embodiment is replaced with the flowchart of common tag detection in FIG. 18 is the same as that of the first embodiment, description thereof is omitted.

  Also, the description of the same part as in FIG. 17 is omitted, and the description will focus on the different part. Note that the comparison table used in the present embodiment includes the tags described in the tag detection table and the lowest image quality score calculated for each tag. Hereinafter, a description will be given using the flowchart of FIG.

  Hereinafter, with reference to the flowchart of FIG. 18, the details of the process (S914) for detecting a common tag performed by the image processing apparatus 100 according to the present embodiment will be described.

  In S1801, the key tag determination unit 304 acquires the maximum score value acquired in S1701 and the value of the corresponding tag T in the comparison table, compares both values, and compares the lowest value other than 0 as the lowest score value. Record on the table.

  In step S1802, the key tag determination unit 304 substitutes the lowest score value of the corresponding tag T in the comparison table into the variable Max. In step S1803, the key tag determination unit 304 compares the lowest score value of the corresponding tag T in the comparison table with the value of the variable Max, and substitutes the larger value into the variable Max. Above, description of each process of the flowchart of FIG. 18 is complete | finished.

  As described above, in the present embodiment, the maximum value of the image quality score is acquired from each image set among the tags that exist in common, and the minimum value is acquired from the plurality of acquired maximum values. Then, among the common tag candidates, the lowest values are compared with each other, and the tag having the maximum value is determined as a key tag. Thus, in the present embodiment, an object having an image with a generally high score can be selected by using the maximum tag among the minimum values for each tag without using the average.

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

  201: Application operation screen 202: Event selection button 203: Creation button 204: Preview area 205: Output button 206: End button 300: Image DB 301: Event type acquisition unit 302: Image set extraction unit , 303: Event DB, 304: Key tag determination unit, 305: Key image selection unit, 306: Non-key image selection unit, 307: Selected image output unit, 308: Creation condition DB

Claims (15)

Determining means for determining the attribute of the key image from the attributes of the subject that exist in common through a plurality of image sets;
Key image selection means for selecting the key image from each image set using the determined attribute;
An image processing apparatus comprising: a non-key image selection unit that selects a non-key image that is an image other than the key image from each of the image sets. An acquisition means for acquiring the event type;
Extraction means for extracting a plurality of image sets corresponding to the event type;
The image processing apparatus according to claim 1, further comprising:   The image processing apparatus according to claim 2, wherein each image set is classified according to the event type and the shooting time of the image.   The determining means acquires a list of attributes of the subject corresponding to the event type, and determines the attributes of the key image from the list of attributes that exist in common through the plurality of image sets. The image processing apparatus according to claim 2.   The image processing apparatus according to claim 4, wherein the determination unit determines an attribute of the key image from the attribute list based on a predetermined priority of the attribute.   The determination means determines the attribute of the key image from the attribute list based on the number of detected images having each attribute included in the attribute list for each image set. The image processing apparatus described.   The image processing apparatus according to claim 4, wherein the determination unit determines an attribute of the key image from the attribute list based on an image quality score of each image included in each image set.   2. The apparatus according to claim 1, further comprising output means for arranging the key image and the non-key image on a page in the album based on a predetermined album creation condition and outputting album creation information. 8. The image processing device according to any one of items 7 to 7.   The creation condition includes information indicating an arrangement position of the key image and the non-key image in the page, and information indicating a size of a display area in the page of the key image and the non-key image. The image processing apparatus according to claim 8, wherein the apparatus is an image processing apparatus.   The image processing apparatus according to claim 8, wherein the creation condition includes setting a display area of the key image to be larger than a display area of the non-key image.   The image processing apparatus according to claim 8, wherein the creation condition includes aligning a position of a display area of the key image for each page of the album. The creation conditions include the number of seconds of the movie, the arrangement position on the time axis for each image, the display time for each image,
The image processing apparatus according to claim 8, wherein the output unit outputs the key image and the non-key image as the movie based on the creation condition.   The image processing according to any one of claims 1 to 12, wherein an image included in each image set is assigned an image ID, a shooting date and time, an event type, a subject attribute, and an image quality score. apparatus. A control method for an image processing apparatus, comprising:
A step of determining an attribute of a key image from attributes of a subject that exist in common through a plurality of image sets;
A step of selecting a key image from each set of images using the determined attribute;
A method for controlling an image processing apparatus, comprising: a step of selecting a non-key image that is an image other than the key image from each of the image sets.   The program for making a computer perform each process of the control method of the image processing apparatus of Claim 14.

Images