JP4457988B2 - Image management apparatus, image management method, and computer program - Google Patents

Description

  The present invention relates to an image management apparatus that manages a plurality of images, and more particularly, to a management name setting process for managing images.

  In general, when a plurality of images taken with an imaging device such as a digital still camera are stored and managed in a personal computer or the like, management is performed by classifying them into folders (groups). More specifically, a new folder is created, a folder name of the folder is determined and input, and an image is stored in this folder. The folder name determined here is preferably common to a plurality of images stored in the folder. This is because, by assigning an appropriate folder name, it becomes easy to search for and view a desired image at a later date.

  Conventionally, various techniques for automatically managing such a plurality of images, in particular, assigning an appropriate folder name without manual operation have been studied. For example, in Patent Document 1 below, a plurality of captured images are divided into groups using position information given to images, a representative image of the group is selected, and a name related to a location given to the representative image is used as a folder name. Techniques for determining are disclosed. According to the image management technique, the name corresponding to the shooting position is used as the folder name, so that an easy-to-understand folder name suitable for the shooting location can be given.

JP 2005-94810 A

  However, with such image management technology, it has been difficult to assign an appropriate name as a folder name. For example, there are a plurality of names given from different viewpoints, such as a country, a prefecture, a city, a town, or a famous place, as names related to the position or place where the image is taken. Therefore, even if the name of the place given to one image is simply used as the folder name, it is not an appropriate folder name common to the images in the folder.

  Furthermore, even when a captured image is a specific event such as a wedding ceremony or athletic meet, or a specific subject such as a flower or animal, the above-mentioned image management technology sets the name of the shooting location as the folder name. Therefore, it was difficult to assign an appropriate folder name.

  SUMMARY An advantage of some aspects of the invention is that it provides an image management apparatus that easily assigns an appropriate management name in consideration of the difficulty of assigning an appropriate management name.

  In view of the above problems, the image management apparatus of the present invention employs the following technique. That is, an image management apparatus for managing a plurality of images, wherein the input means inputs a plurality of pieces of image data having image keywords expressing the shooting location information of the images and / or the image contents of the images in a predetermined format. And using the image keywords of the plurality of input image data, refer to a database stored by assigning a plurality of keywords in advance and managing the plurality of image data with keywords related to the image keywords. The gist of the present invention is that a management means for associating the plurality of image data with the management name is provided.

  The image management method of the present invention is an image management method for managing a plurality of images, and includes a plurality of image keywords that represent shooting location information of the images and / or image contents of the images in a predetermined format. The image keyword of the plurality of input image data is input to the database stored in advance by giving a hierarchical structure to the plurality of keywords, and the keywords related to the image keyword are selected. The gist is to associate the plurality of image data with the management name.

  According to the image management apparatus and the image management method of the present invention, the management name is set by using the image keyword assigned to each image data and referring to the database storing the hierarchical keyword. Further, the set management name is a name related to the image keyword. Therefore, an appropriate management name related to the image keyword can be easily assigned. Further, since image data can be managed by a name associated with an image keyword, it is possible to easily manage a plurality of images.

  The keyword of the database of the image management apparatus having the above configuration is formed in a hierarchical structure in which the upper layer keyword includes the lower layer keyword, and the management means refers to the database and the predetermined number or more of the images A keyword extraction unit that extracts a keyword common to the keywords from the database, and at least one keyword is selected from the extracted keywords, and the selected keyword is assigned to the folder that stores the plurality of image data. It is good also as a thing provided with the folder name determination means determined as a folder name.

  According to such an image management apparatus, keywords common to predetermined or more image keywords are extracted from a hierarchical database, and a folder name is determined. Therefore, an appropriate folder name common to many of the image data in the folder can be given. By assigning an appropriate folder name, image management such as searching for a desired image can be facilitated.

  An input unit of the image management apparatus having the above configuration includes a group classification unit that classifies the plurality of input image data to generate a group, and the management unit includes one or more images of each of the generated groups. A folder name may be determined for each folder storing data.

  According to such an image management apparatus, it is possible to classify a plurality of input image data into several groups and set a folder name for each folder. For example, even when there are a plurality of image data that should be classified into several groups, these image data are input and grouped, and an appropriate folder name is set for each folder. Can do.

  The plurality of image data input by the image management apparatus having the above-described configuration includes at least one of geographical position information of longitude and latitude or shooting date / time information, and the group classification unit is based on the information. The plurality of image data may be classified into groups.

  According to such an image management apparatus, a plurality of image data can be classified into several groups relatively easily.

  The keyword extraction means of the image management apparatus having the above configuration includes a counter for each keyword provided corresponding to the keyword of each level stored in the database, and to which level of keyword the image keyword corresponds. And the process of counting the counter of the keyword corresponding to the upper layer of the searched keyword is executed for the image keywords of the plurality of image data. And means for extracting a keyword having a counter value equal to or greater than a predetermined value as a keyword common to the image keyword equal to or greater than the predetermined value.

  According to such an image management apparatus, a keyword common to a predetermined number or more of image keywords is extracted based on a counter value for each keyword of each hierarchy stored in the database. That is, by comparing the counter value with the threshold value, it is determined whether or not to extract as a common keyword. Therefore, it is easy to determine a keyword to be extracted.

  The keyword extraction unit of the image management apparatus having the above configuration may exclude image data having a preset image keyword from the processing target.

  According to such an image management apparatus, a predetermined image keyword is set in advance, and image data corresponding to this is excluded from the processing targets. Examples of the image keyword set in advance include a home position and location, a pet name, and the like. By removing image data including such image keywords from a plurality of image data to be used for determining a folder name, an appropriate folder name common to many of the plurality of image data can be given.

  The folder name determination unit of the image management apparatus having the above configuration may select a keyword corresponding to a lower layer among the extracted keywords when there are a plurality of extracted keywords.

  According to such an image management apparatus, a keyword corresponding to a lower layer is selected from a plurality of extracted keywords. Therefore, it is common to most of the image data in the folder and a narrowly defined keyword can be assigned as the folder name. That is, a specific keyword can be set as the folder name, which can be easily understood.

  When there are a plurality of keywords selected as corresponding to the lower layer, the folder name determination unit of the image management apparatus having the above configuration may use the keywords as folder names in parallel.

  According to such an image management apparatus, when there are a plurality of selected keywords, the folder names are set in parallel. For example, when two keywords are selected, the upper layer keyword including the two keywords is not used as the folder name, but the two keywords are arranged side by side as the folder name. In this way, a specific and easy-to-understand folder name can be assigned.

  The shooting location information of the image management apparatus having the above configuration is the place name of the location where the image was shot, and the image content is information on the subject at the time of shooting the image and / or event information at the time of shooting. Yes, the input means may handle the place name and / or information representing the information in a text format as the image keyword. In this way, an appropriate folder name corresponding to the contents of the image data, such as a place name, a subject name, and an event name, can be assigned.

  In the image data of the image management apparatus having the above-described configuration, geographical position information expressed by longitude and latitude is attached as the shooting location information, and the input unit is configured based on the geographical position information, It is good also as a thing provided with the conversion means which calculates | requires the place name of the place which image | photographed the said image as said image keyword. By doing so, it is possible to easily perform processing on image data having position information of longitude and latitude.

  The management means of the image management apparatus having the above-described configuration may include means for writing a keyword having the management name as additional information of the plurality of image data. By doing so, a management name is given to the image data itself, and processing such as image search becomes easy.

  The management means of the image management apparatus having the above configuration selects a predetermined number of image data from image data including a keyword with the management name as the image keyword, and uses the selected image data as a thumbnail image. The management name may be associated with the plurality of image data. By doing so, for example, a folder name as a management name can be displayed and a thumbnail image can be displayed, and it is easy to guess image data stored in the folder.

  The present invention can also be implemented as a computer program and a medium recording the computer program. As the recording medium, various computer-readable media such as a flexible disk, a CD-ROM, a DVD-ROM / RAM, a magneto-optical disk, a memory card, and a hard disk can be used.

Hereinafter, embodiments of the present invention will be described in the following order based on examples.
A. Image management device:
B. Image management device structure:
C. Classification process:
C-1. Folder name determination process:
C-2. Concrete example:
D. effect:
E. Variations:

A. Image management device:
FIG. 1 is an explanatory diagram showing an image management apparatus 10 as an embodiment of the present invention. The image management apparatus 10 classifies the image data D into a predetermined group (folder) based on position information and keywords added to the plurality of image data D, and assigns an appropriate folder name to the classified folder. Thus, the apparatus manages a plurality of image data D in groups. The image data D handled in this embodiment is a photographed image photographed by an imaging device such as a digital still camera. In addition to the location information of the photographing location when the image is photographed, the place name of the photographing location is used as a keyword in text format The represented information is also added to the image data D.

  As shown in the figure, the image management apparatus 10 includes a group classification unit 14, a keyword extraction unit 16, a folder name determination unit 18, and the like, and is connected to a predetermined database 50.

  The group classification unit 14 receives a plurality of image data D photographed by the imaging device and classifies them into groups. Specifically, the group classification unit 14 sequentially inputs a plurality of image data D, and performs grouping based on position information for each input image data D.

  The keyword extraction unit 16 extracts a predetermined keyword from the database 50 based on the keywords for each of the plurality of image data D classified into one group by the group classification unit 14. The database 50 is provided with a keyword group of place names in a hierarchical structure (tree form) in which elements in the upper layer include elements in the lower layer. The keyword extraction unit 16 compares the place name of the shooting place for each image data D with the place name keyword group in the database 50, and extracts a keyword that matches a predetermined condition.

  The folder name determination unit 18 selects a keyword corresponding to the lowest layer among keywords extracted satisfying a predetermined condition, and outputs the keyword as a group folder name.

  The image management apparatus 10 having such processing units can be configured as hardware or software. Hereinafter, the image management apparatus 10 in which each of these processing units is configured as a program module for image management processing, which is software, and this program is installed in a personal computer will be described.

  In the following description, a process of classifying a plurality of image data D into a group is also included. However, when a plurality of image data to be processed is already a group, the keyword extracting unit 16 The image management apparatus 10 is configured by a processing program in the folder name determination unit 18.

B. Image management device structure:
FIG. 2 is a schematic structural diagram of an image management apparatus 10 provided with the image management processing program of the present invention as software. As shown in the figure, the image management device 10 captures a predetermined image to be processed, inputs the digital still camera 20 storing the image data, and the image data stored in the digital still camera 20, and performs predetermined processing. A personal computer 30 (hereinafter referred to as a PC 30) to be applied, a display 40 for displaying an image processed by the PC 30, a keyboard 41 and a mouse 42 as a user interface, and the like.

  The digital still camera 20 includes a CCD as an image sensor, an image processing circuit that performs predetermined image processing on image data captured via the CCD, a memory card 24 that stores the image data, and the like, and is an aspect of wireless communication. A GPS function for storing geographical position information of the digital still camera 20 using a certain GPS (Global Positioning Systems) communication is provided.

  Position information by GPS is added to the image data captured by the digital still camera 20. In general, the data structure of an image handled by a digital still camera is a so-called Exif format. Based on image data in JPEG format, predetermined image additional information such as shooting information when shooting image data, thumbnail images, etc. Embedded in a format that conforms to the rules. As the image additional information, various information such as the shooting date and time by the internal clock of the digital still camera 20, the exposure time at the time of shooting, the aperture value, the shutter speed, the ISO sensitivity, and the white balance are stored. In addition to this, as one of the image additional information, position information of latitude and longitude by GPS and a place name keyword of the shooting location are included.

  In addition to this configuration, the digital still camera 20 according to the present embodiment includes a character input key 20a. When the user inputs a place name of the shooting place using the character input key 20a, a place name keyword is added to the image data. Is done. There are various methods for assigning a place name keyword. For example, a place name list may be provided in a hierarchy on the digital still camera 20, and a place name keyword may be assigned to the image data by a user selecting a place name, or the digital still camera 20 may be connected to the Internet. A connection function may be provided, and a corresponding place name may be acquired from position information obtained by GPS via the Internet, and this may be assigned to image data. Further, the digital still camera 20 may be provided with a voice recognition function, and a place name keyword may be given to the image data by voice recognition of the place name issued by the user. In the case of voice recognition, it is possible to cope with matching within a limited range of “place name”, and therefore a place name keyword can be given to image data relatively easily.

  Using such image data as a processing target, a plurality of pieces of image data are grouped using position information of longitude and latitude by GPS.

  If the image data to be handled is provided with position information and keywords, instead of obtaining from the digital still camera 20, the digital video camera 21, the camera-equipped mobile phone 23, the memory card 24, and the hard disk 25 are used. For example, acquisition from an imaging device or a recording medium may be used.

  The PC 30 includes a CPU 31, a ROM 32, a RAM 33, a hard disk 34, an I / F circuit unit 35, and the like, and each device is connected by an internal bus. The I / F circuit unit 35 is connected to the digital still camera 20, the display 40, the keyboard 41, and the mouse 42, and functions as an interface between these external devices and the PC 30.

  A predetermined operation system is stored in the ROM 32, and when the PC 30 is turned on, the CPU 31 reads and starts the operation system OS in the ROM 32.

  Various application programs that operate on the operating system OS are installed in the hard disk 34. In this embodiment, as an image management process, a classification processing program is installed in the hard disk 34 that takes in a plurality of image data from the digital still camera 20, classifies them, and assigns folder names to the classified folders. The CPU 31 receives a user operation instruction via the keyboard 41 and the mouse 42, reads out this program, develops it on the RAM 33, and executes it. In this embodiment, the hierarchical database 50 shown in FIG. 1 is also installed in the hard disk 34.

C. Classification process:
FIG. 3 is a flowchart of a classification process for classifying a plurality of image data. This process is executed by the CPU 31 of the PC 30 after connecting the digital still camera 20 storing a plurality of image data to the PC 30 and instructing the user to take in the image data from the digital still camera 20.

  When the process is started, the CPU 31 sequentially captures the image data in the digital still camera 20 in the order of shooting date and time, and groups the plurality of image data from the position information of the image data corresponding to the front and rear in the order of capturing (step) S300).

  Specifically, every time the image data is captured, the moving distance of the shooting location with the image data captured immediately before is calculated, and when the moving distance is more than a predetermined distance, the image data captured after that is The group is different from the image data that has already been captured. In calculating the moving distance, position information based on longitude and latitude given to each image data is used.

  Subsequently, a folder name determination process for determining a folder name for one grouped group is executed (step S320). In the folder name determination process, a place name keyword added to each image data in one group is compared with a keyword group in the database 50, and a keyword including image data of a predetermined ratio or more in the group is extracted from the database 50. To do. When there are a plurality of extracted keywords, the keyword corresponding to the lowest layer is selected, and this keyword is determined as the folder name of the group. Details of the folder name determination process will be described later.

  After determining one group and a folder name to be given to the group, the CPU 31 writes image data to the hard disk 34 and instructs the operation system OS to write data (step S330). The operation system OS includes a file allocation table for managing various files stored in the hard disk 34 in the PC 30. The CPU 31 instructs to write the correspondence between the determined folder name and the image data belonging to the folder (group) in the file allocation table.

  Subsequently, the CPU 31 determines whether there is an unprocessed group (step S340). If it is determined in step S340 that there is an unprocessed group (Yes), the process returns to step S320, and a series of processes is repeated. On the other hand, if it is determined in step S340 that there is no unprocessed group (No), the series of classification processes ends.

  By such classification processing, a plurality of image data stored in the hard disk 34 is managed as belonging to a folder with a predetermined folder name.

C-1. Folder name determination process:
FIG. 4 is a flowchart of folder name determination processing for determining a folder name. As described above, this process is a process executed by the CPU 31 after grouping a plurality of image data.

  When the process is started, the CPU 31 acquires a keyword assigned to one image data belonging to one group (step S500). As described above, the image data handled in the present embodiment is preliminarily provided with “location name” (for example, “XX prefecture”, “ΔΔ city”) of the shooting location as a keyword. There are various keywords such as “place name” which is information on the shooting location, information on subjects such as “flower” and “dog”, and event information such as “athletic meet” and “wedding”. In the following description, a keyword added to image data as including such various keywords is referred to as an image keyword.

  Even if an image keyword is not added to the image data, it can be handled as long as position information expressed by the longitude and latitude of one image data is added. For example, GPS location information is input, a place name of a shooting location is specified from longitude and latitude information using a predetermined table, and this is used as an image keyword (see FIG. 11). A table associating such longitude and latitude with the place name of the shooting location can be stored in the hard disk 34 in advance.

  After acquiring the image keyword assigned to one image data, the CPU 31 searches the hierarchical keyword corresponding to the acquired image keyword using the database 50 (step S510).

  As described above, the database 50 according to the present embodiment includes a keyword group of place names having a hierarchical structure (tree shape) in which the upper layer element includes the lower layer element. FIG. 5 is an explanatory diagram showing an example of a hierarchical keyword group provided in the database 50. As shown in the figure, the keyword group includes “town name D1,” “town name D2,” “city name C1,” “city name C1,” “city name C2,” “city name C3”. “Country name B1”, “Prefecture name B1”, “Prefecture name B2”, etc., “Country name A1”, etc. "Is the hierarchical structure (tree shape) with the highest layer. Each of the “town name D1”, “town name D2”,... “Country name A1” is called a hierarchical keyword here.

  That is, in the process of step S510, the image keyword (for example, “ΔΔ city”) is compared with the keyword group in the database 50 to search for the corresponding hierarchical keyword (for example, “city name C1”).

  Subsequently, the CPU 31 determines whether or not the hierarchical keyword counter has been counted in correspondence with the image keyword of the image data to be processed, that is, whether or not it has already been counted (step S520). Each hierarchical keyword constituting the keyword group in the database 50 is provided with a counter, and the hierarchical keyword corresponding to the image keyword to be processed is counted up as will be described later. By counting in this way, it is evaluated which hierarchy of image keywords of one image data is appropriate to be subordinate to. Note that the count value of this counter is used when finally determining the folder name.

  If it is determined in step S520 that the count has not been completed (No), the hierarchical keyword counter searched in step S510 is incremented by one (step S530). Subsequently, if the upper layer hierarchical keyword to which the searched hierarchical keyword belongs is not counted, the upper layer hierarchical keyword counter is incremented by one (step S540), and the process proceeds to the next step S550. In this series of steps, not only the searched hierarchical keyword but also the upper hierarchical keyword including the hierarchical keyword is counted.

  On the other hand, if it is determined in step S520 that the count has been completed (Yes), the process proceeds to the determination process in step S550.

  In step S550, CPU 31 determines whether or not the evaluation has been completed for all image keywords assigned to one image data. That is, in this step, it is assumed that a plurality of image keywords are assigned to one image data.

  If it is determined in step S550 that evaluation for all image keywords has not been completed (No), the process returns to step S510, and a series of evaluation processes (counting) is repeated for image keywords that have not been evaluated. On the other hand, if it is determined in step S550 that the evaluation for all image keywords has been completed (Yes), the process proceeds to step S560.

  In step S560, CPU 31 determines whether or not the evaluation of all image data of one group has been completed. If it is determined in step S560 that the evaluation processing has not been completed for all image data (No), the process returns to step S500, and a series of evaluation processing is performed for the next processing target image data belonging to one group. repeat. On the other hand, if it is determined in step S500 that the evaluation has been completed for all the image data (Yes), the process proceeds to step S570.

  The evaluation process so far will be specifically described. For example, when there are two image keywords “△△ city” and “XX prefecture” for one image data to be processed, first, a hierarchical keyword is searched for “△△ city”, and “city name” It is assumed that “C1” is searched (step S510). In this case, one counter value of “city name C1” is counted, and “prefecture name B1” and “country name A1” corresponding to the upper layer are also counted (steps S530 and S540).

  Subsequently, a hierarchical keyword is searched for “XX prefecture” as the next image keyword, and “prefecture name B1” is searched (step S510). Here, when the count value of “prefecture name B1” is counted, the count value of “prefecture name B1” has already been counted in correspondence with the processing for “ΔΔ city”. In this case, the process goes from step S520 to step S550 to determine whether or not there is a next image keyword. In this example, since all the image keywords have been evaluated, the process proceeds to step S560.

  On the other hand, when there are two image keywords “△△ city” and “XX prefecture”, first, a hierarchical keyword is searched for “XX prefecture”, and “prefecture name B1” is searched. (Step S510). In this case, one count value of “prefecture name B1” is counted, and “country name A1” corresponding to the upper layer is also counted (steps S530 and S540).

  Subsequently, a hierarchical keyword is searched for “ΔΔ city” as the next image keyword, and “city name C1” is searched (step S510). Here, “city name C1” is not counted in correspondence with the processing for “prefecture XX”, and therefore the process proceeds to step S530 and the counter value of “city name C1” is counted by one. To do. In step S540, “prefecture name B1” and “country name A1” corresponding to the upper layer of “city name C1” are also counted. Since these upper layers are already counted, they are counted here. Instead, the process proceeds to step S560.

  That is, by a series of evaluation processes so far, even if there are a plurality of image keywords assigned to the image data, the corresponding hierarchical keywords are assigned to all the image data belonging to one group without performing overlapping counting. Search for. In other words, each image data in one group is assigned to one of the hierarchical keywords using an image keyword, and the number of assigned image data is displayed as a counter value. In order to eliminate such a duplicate count, a flag indicating that the processing has been completed is set in each step in relation to the image data and the image keyword. The CPU 31 determines whether the processing has been completed by checking the flag.

  Thus, after evaluating which hierarchical keyword the image data belongs to, the CPU 31 extracts a hierarchical keyword whose counter value is equal to or greater than a preset threshold value (step S570).

  Subsequently, when there are a plurality of extracted hierarchical keywords, the CPU 31 selects a hierarchical keyword corresponding to the lowest layer as a representative keyword (step S580), and determines the representative keyword as a folder name (step S590). ). That is, this representative keyword is given to a folder as a name (management name) for managing a plurality of image data. In this way, a series of folder name determination processing is completed, and thereafter, the processing shifts to S330 shown in FIG.

  For example, when two hierarchical keywords “prefecture name B1” and “country name A1” are extracted as hierarchical keywords whose counter value is equal to or greater than the threshold value in step S570, “prefectural name B1 corresponding to the lowest layer among them” is extracted. "Is determined as the folder name. When there is one hierarchical keyword that is equal to or greater than the threshold, it is determined as it is as the folder name.

C-2. Concrete example:
An application example of such folder name determination processing will be specifically described. FIG. 6 is an explanatory diagram showing, on a map, shooting locations where a plurality of images to be processed are shot. The “x” mark on the illustrated map indicates the place where the image was shot, and the numbers attached to the side of the “x” mark indicate the shooting order. In this example, it is assumed that 20 pieces of image data are made into one group.

  As shown in the figure, 7 in "town name D1", 8 in "town name D2", 1 in "city name C2", 2 in "city name C3", and in "prefecture name B2" Two images including “Home” are taken respectively. That is, most of the 20 pieces of image data are taken with “town name D1” and “town name D2”.

  When these 20 pieces of image data are evaluated according to the folder name determination process, as shown in FIG. 7, “prefecture name B1”, “prefecture name B2”, “city name C1”, “city name C2”, “city name” The counters of the hierarchical keywords “C3”, “town name D1”, and “town name D2” are counted. In addition, although explanation is omitted here, the counter of “country name”, which is actually the upper layer of “prefecture name B1”, is also counted.

  Here, assuming that the threshold for extracting a hierarchical keyword is 10, “prefecture name B1” having a counter value of 18 and “city name C1” having a counter value of 15 are extracted as hierarchical keywords. Of these two hierarchical keywords, “city name C1” corresponding to the lowest layer is determined as the folder name. By applying the folder name determination process, a hierarchical keyword that is common to most of a plurality of image data in one group is selected, so that an appropriate level name can be given as a name representing the group. .

  There may be a plurality of hierarchical keywords assigned as folder names. For example, in the specific example shown in FIG. 7, when the threshold for extracting the hierarchical keyword is 7, there are four “prefecture name B1”, “city name C1”, “town name D1”, and “town name D2”. Hierarchy keywords are extracted, and the lowest ones are “town name D1” and “town name D2”. In this case, the image data belonging to each of the hierarchical keywords is 7/20 and 8/20, and image data with a substantially similar ratio belongs. If the conditions such as the same level and the same ratio, the condition that the number of characters as the level keyword is a predetermined number, and the like are satisfied, a plurality of level keywords may be set in parallel as the folder name. In this example, “town name D1 town name D2” is the folder name. In this way, a specific and easy-to-understand folder name can be assigned.

D. effect:
As described above, according to the classification process including the folder name determination process, a plurality of image data is grouped and a folder name is automatically given to each group (folder). Since this folder name is set by evaluating image keywords of image data belonging to the folder, an appropriate name common to many of the image data in the folder can be determined as the folder name. Since image data is stored in a folder with an appropriate folder name in this way, image management such as searching for a desired image can be facilitated.

  For the evaluation of the image keyword, a database having a hierarchical keyword group is used, and a keyword corresponding to the lower layer is determined as a folder name. That is, among hierarchical keywords extracted as common to many image data, a narrow (specific) hierarchical keyword is used as the folder name. Therefore, it is possible to set a specific and easy-to-understand name as the folder name that is common to many image data.

  Furthermore, a counter provided for each hierarchical keyword is used to determine the hierarchical keyword to be extracted. Therefore, it is possible to easily determine how much of the image data is a hierarchical keyword only by looking at the count value of the counter.

  Further, according to the classification process of this embodiment, a plurality of image data are grouped based on the position information. Originally, even when multiple image data that should be classified into several groups are mixed, these image data can be input and classified, and an appropriate folder name can be assigned to each group. it can.

  In the folder name determination process of the present embodiment, the folder name is determined using the hierarchical keyword group of place names shown in FIG. 5, but the keyword group stored in the database 50 is not limited to this.

  For example, instead of the place name keyword group, or together with the place name keyword group, a keyword group related to animals shown in FIG. 8 may be provided. As shown in FIG. 8, this keyword group is not hierarchically structured based on zoological classification such as mammals and reptiles, but is hierarchically formed based on words that are easy to use on a daily basis, such as wild animals and pets. It is what is done. In other words, it is a keyword group (thesaurus) specialized for the folder name of the captured image that is highly likely to be set by the user. By using the thesaurus database 50 specialized for daily human behavior such as places, events, and objects, it is possible to search for an appropriate keyword as a folder name. By providing such a keyword group, not only the shooting location (place name) but also the name of the event or subject can be set as the folder name, and an appropriate folder name corresponding to the image contents in the group can be given. .

  In forming such a keyword group, “other” hierarchical keywords can be provided independently as shown in FIG. When the hierarchical keyword cannot be searched from the image keyword of the image data, the image data is counted as belonging to “others”. As a result of executing the evaluation process on all the image data in one group in this way, when there is image data belonging to “others” more than a predetermined threshold, it is assumed that there is no appropriate folder name, and the user is prompted to input the folder name. It should be.

  When the database 50 is configured by including a plurality of such keyword groups (thesaurus), first, it is determined to which thesaurus most image data (image keywords) corresponds to all image data of one group. As a result of the determination, one thesaurus may be selected and the folder name may be determined using this.

  Further, the place names shown in FIG. 5 are simply a hierarchical structure based on prefectures, municipalities, etc., but the names of sightseeing spots may be used as the place names. In this way, a friendly folder name can be set.

  In this embodiment, as shown in FIG. 6, a photographed image at home is also subject to evaluation. However, for example, specific position information such as a home position, a frequently visited place, or a pet, a keyword May be set in advance, and image data corresponding to this may be excluded from the evaluation target. By performing such setting, the ratio of the common hierarchical keywords in the remaining image data to be processed increases, and a more appropriate folder name can be determined.

E. Variations:
The embodiments of the present invention have been described above. However, the present invention is not limited to these embodiments, and can of course be implemented in various forms without departing from the spirit of the present invention. is there.

  In the present embodiment, the image data assigned with the folder name by the classification process and classified in the predetermined folder is stored in the hard disk 34. However, the representative keyword selected by the folder name determination process is stored in each image data. It may be given (output) to.

  For example, as shown in the structure of the image data in FIG. 9, the Exif format image data targeted in this embodiment includes image additional information such as shooting information and GPS information. This image additional information includes an area (maker note) that can be freely used by the user. By writing a representative keyword as a management name in such an area and associating the image data with the management name, it can be effectively used in various image processing and image management such as retrieval and editing of captured images.

  Also, a plurality of image data having the same representative keyword as the selected representative keyword is selected, and several pieces of image data are extracted based on the selected criteria, and are associated with the folder as thumbnail images of the folder. It may be stored and displayed on the display 40 in a predetermined format.

  FIG. 10 shows a display mode in which the folder into which the image data is classified is displayed on the display 40. As shown in the figure, on the display, a plurality of folders assigned predetermined folder names by the classification process of this embodiment are displayed as icons K1, K2, K3,. On the icons K1, K2, K3,..., Three image data extracted based on a predetermined standard are displayed as thumbnails. By displaying the folder in this manner, it is possible to easily recognize what kind of image data is included in one folder.

  Note that the predetermined criteria for extracting image data for thumbnail display include various existing image evaluation methods such as evaluating the feature amount of image data and the relative importance of images in a group. It should be used. Alternatively, the shooting date and time in the folder may be in order from the earliest.

  In the classification process of this embodiment, the movement distance is obtained from the position information of the longitude and latitude of the GPS, and the grouping is performed. However, the grouping may be performed simply based on the shooting date and time. By using the shooting date and time, grouping becomes easy.

  In this embodiment, the place name is assigned as an image keyword. However, for example, when an object unique to the place such as a building is an image keyword, the place name may be temporarily converted into an image keyword of the place name. good.

  For example, the conversion table shown in FIG. This conversion table is a table in which a building name unique to the place is associated with position information (longitude, latitude) and place name information of the place. If the image keyword assigned to one image data is “Tokyo Tower”, it can be converted into place name information such as “Tokyo” or “Minato-ku” using the conversion table. By converting into place name information in this way, the place name keyword group shown in FIG. 5 can be used as it is.

  In the present embodiment, the database 50 has been described as being provided in the hard disk 34 of the PC 30, but it is also possible to separately provide a database as a server on the Internet and acquire hierarchical keywords via the Internet. In this case, the I / F circuit 35 of the PC 30 may have a function for connecting to the Internet line.

  The image management apparatus of the present embodiment is configured by software, but of course, may be configured by hardware. Further, not only the PC 30 but also a digital still camera or the like may be added with a classification processing function to form an image management apparatus.

It is explanatory drawing which shows the image management apparatus as one Example of this invention. It is a schematic structure diagram of an image management apparatus provided with an image management processing program as software. It is a flowchart of the classification process which classifies a plurality of image data. It is a flowchart of the folder name determination process which determines a folder name. It is explanatory drawing which shows an example of a hierarchical keyword group. It is explanatory drawing which showed the imaging | photography place which image | photographed the several image on the map. It is explanatory drawing of the specific example of a folder name determination process. It is explanatory drawing which shows an example of a hierarchical keyword group. It is explanatory drawing which shows the structure of image data. This is a display mode in which a folder into which image data is classified is displayed on a display. It is explanatory drawing of the conversion table converted into the image keyword of a place name.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Image management apparatus 14 ... Group classification | category part 16 ... Keyword extraction part 18 ... Folder name determination part 20 ... Digital still camera 21 ... Digital video camera 23 ... Mobile with camera Telephone 24 ... Memory card 25 ... Hard disk 30 ... Personal computer 31 ... CPU
32 ... ROM
33 ... RAM
34 ... Hard disk 35 ... I / F circuit part 40 ... Display 41 ... Keyboard 42 ... Mouse 50 ... Database

Claims (13)

An image management apparatus for managing a plurality of images,
Input means for inputting a plurality of image data including an image keyword expressing the shooting location information of the image and / or the image content of the image in a predetermined format;
Management for managing a plurality of keywords related to the image keyword by referring to a database in which a plurality of keywords are given a hierarchical structure in advance using the image keywords of the input image data A management means for associating the plurality of image data with the management name as a name ,
The keywords in the database are formed in a hierarchical structure in which upper-layer keywords contain lower-layer keywords,
The management means includes
A keyword extracting means for referring to the database and extracting a keyword common to the image keywords of a predetermined value or more from the database;
Folder name determining means for selecting at least one keyword from the extracted keywords, and determining the selected keyword as a folder name to be given to a folder storing the plurality of image data;
The keyword extracting means includes
A counter for each keyword provided corresponding to each level keyword stored in the database;
A process of searching for which level of the keyword the image keyword corresponds, counting the counter of the searched keyword, and counting the counter of the keyword corresponding to the upper layer of the searched keyword, Means for executing image keywords of the plurality of image data;
Means for extracting a keyword whose counter value is a predetermined value or more as a keyword common to the predetermined or more image keywords;
An image management apparatus comprising: The image management apparatus according to claim 1 ,
The input means includes group classification means for generating a group by classifying the input plurality of image data,
The image management apparatus, wherein the management unit determines a folder name for each folder storing one or more image data of each of the generated groups. The image management apparatus according to claim 2 ,
The plurality of input image data includes at least one of geographical position information of longitude and latitude, or information of photographing date and time,
The group classification unit is an image management apparatus that classifies the plurality of image data into groups based on the information. The image management apparatus according to claim 1 ,
The keyword extraction unit is an image management apparatus that excludes image data having a preset image keyword from a processing target. The image management apparatus according to any one of claims 1 to 4 ,
The folder name determining means, when there are a plurality of the extracted keywords, an image management apparatus for selecting a keyword corresponding to a lower layer among the keywords. The image management apparatus according to claim 5 ,
When there are a plurality of keywords selected as corresponding to the lower layer, the folder name determination unit is an image management apparatus that uses the keywords as a folder name in parallel. The image management apparatus according to any one of claims 1 to 6 ,
The shooting location information is a place name where the image was taken,
The image content is information on a subject at the time of shooting the image and / or information on an event at the time of shooting,
The image management apparatus which handles the information which expressed the place name and / or the information in a text format as the image keyword. The image management apparatus according to any one of claims 1 to 6 ,
In the image data, as the shooting location information, geographical position information represented by longitude and latitude is attached,
The image management apparatus includes a conversion unit that obtains a place name of a place where the image is captured as the image keyword from the geographical position information. The image management apparatus according to any one of claims 1 to 8 ,
The image management apparatus includes a means for writing a keyword having the management name as additional information of the plurality of image data, respectively. An image management apparatus according to any one of claims 1 to 9 ,
The management means selects a predetermined number of image data from image data having a keyword with the management name as the image keyword, and uses the selected image data as a thumbnail image together with the management name. An image management device associated with image data. An image management method for managing a plurality of images,
Computer, an input step of the image content of the captured location information and / or the image of the image, inputting a plurality of image data having an image keyword expressed in a predetermined format,
The computer uses the image keywords of the inputted plurality of image data to refer to a database stored in advance by giving a hierarchical structure to the plurality of keywords, and the keywords related to the image keyword are referred to as the plurality of image data. And a management step for associating the plurality of image data with the management name ,
The keywords of the database are formed in a hierarchical structure in which upper-level keywords include lower-level keywords, and counters are associated with the keywords of each level,
The management process includes
A keyword extracting step in which the computer refers to the database and extracts a keyword common to the image keywords of a predetermined value or more from the database;
A folder name determining step in which the computer selects at least one keyword from the extracted keywords, and determines the selected keyword as a folder name to be given to a folder storing the plurality of image data. And
The keyword extraction step includes
The computer searches to which hierarchy keyword the image keyword corresponds, and counts the counter of the searched keyword, and also counts the counter of the keyword corresponding to the upper layer of the searched keyword. Performing a process to target image keywords of the plurality of image data;
The computer extracting a keyword whose counter value is equal to or greater than a predetermined value as a keyword common to the predetermined or greater image keywords;
An image management method comprising: A computer program for controlling an image management apparatus that manages a plurality of images,
An input function in which the image management device inputs a plurality of pieces of image data including image keywords expressing the shooting location information of the image and / or the image content of the image in a predetermined format;
The image management apparatus refers to a database stored in advance by giving a hierarchical structure to a plurality of keywords using the image keywords of the plurality of input image data, and selects a keyword having a relation to the image keywords. A management name for managing image data, and a management function for associating the plurality of image data and the management name with each other .
The management function is
The image management device refers to the database, and extracts a keyword common to the image keywords of a predetermined value or more from the database;
A folder name determination function in which the image management device selects at least one keyword from the extracted keywords, and determines the selected keyword as a folder name to be given to a folder storing the plurality of image data; Is realized in the image management device,
The keywords of the database are formed in a hierarchical structure in which upper-level keywords include lower-level keywords, and counters are associated with the keywords of each level,
The keyword extraction function is
The image management device searches for which level of the keyword the image keyword corresponds to, and counts the counter of the searched keyword, and for the counter of the keyword corresponding to the upper layer of the searched keyword A function of executing a process for counting image keywords of the plurality of image data,
A computer program for causing the image management apparatus to realize a function of the image management apparatus to extract a keyword having a counter value equal to or larger than a predetermined value as a keyword common to the predetermined or larger image keyword . A recording medium in which the computer program according to claim 12 is recorded in a computer-readable manner.

Images