JP2015026105A - Image evaluation server - Google Patents

Abstract

PROBLEM TO BE SOLVED: To appropriately evaluate an image.SOLUTION: An image evaluation server comprises: a communication unit 12 that acquires image data; a keyword generation unit 16 that generates a keyword corresponding to at least one subject image included in the image data; an evaluation search unit 13 that determines an evaluation score in acquiring the image data by the communication unit for each keyword generated by the keyword generation unit; and an information DB 22 that records an evaluation score for each keyword determined by the evaluation search unit in association with the image data.

Description

  The present invention relates to an image evaluation server.

  2. Description of the Related Art Conventionally, there has been known an evaluation device that weights an evaluation performed by a user who is considered to have a high degree of skill in photography, and determines an evaluation of a photograph (see, for example, Patent Document 1).

JP 2009-294767 A

  However, this evaluation apparatus has a problem that an objective evaluation value cannot be determined for a subject image included in a photograph, and thus an image cannot be evaluated accurately.

  An object of the present invention is to provide an image evaluation server that can accurately evaluate an image.

  The image evaluation server of the present invention includes an image acquisition unit that acquires image data, a keyword acquisition unit that acquires a keyword corresponding to at least one subject image included in the image data, and the keyword acquired by the keyword acquisition unit. A determination unit that determines an evaluation score when the image data is acquired by each image acquisition unit, and a recording unit that records the evaluation score for each keyword determined by the determination unit in association with the image data; It is characterized by providing.

  According to the image evaluation server of the present invention, an image can be accurately evaluated.

It is a block diagram which shows the structure of the image evaluation system which concerns on embodiment. It is a block diagram which shows the structure of the image evaluation server which concerns on embodiment. It is a block diagram which shows the structure of the terminal which concerns on embodiment. It is a figure which shows the site screen of the image evaluation site provided by the image evaluation server which concerns on embodiment. It is a figure which shows the structure of the link table memorize | stored in the information database of the image evaluation server which concerns on embodiment. It is a flowchart which shows the process which submits an image to the image evaluation site provided by the image evaluation server which concerns on embodiment. It is a flowchart which shows the image evaluation process in the image evaluation server which concerns on embodiment. It is a figure which shows the display of the site screen provided by the image evaluation server which concerns on embodiment. It is a figure which shows the map displayed on the site screen provided by the image evaluation server which concerns on embodiment. It is a figure which shows the map displayed on the site screen provided by the image evaluation server which concerns on embodiment. It is a figure which shows the site screen of the image evaluation site provided by the image evaluation server which concerns on embodiment.

  Hereinafter, an image evaluation system according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a configuration of an image evaluation system 2 according to the embodiment. The image evaluation system 2 includes an image evaluation server 4 that performs image evaluation processing, and the image evaluation server 4 can communicate with various terminals 8 and an external search site 9 via a network 6.

  Here, the terminal 8 includes a personal computer that can store image data in addition to an imaging device that can capture an image, such as a digital camera, a smartphone, a tablet terminal, and a mobile phone. Each terminal 8 has a built-in communication device that can be connected to the image evaluation server 4 or is mounted as an external device.

  The external search site 9 includes a plurality of existing image search sites such as a search site A, a search site B,.

  FIG. 2 is a block diagram illustrating a configuration of the image evaluation server 4 according to the embodiment. The image evaluation server 4 is a server that displays an image in association with position information and provides an image evaluation site (see FIG. 4) that receives an image evaluation from a communicable terminal 8. CPU10 which controls this is provided. The CPU 10 includes a communication unit 12 that communicates with the terminal 8 via the network 6, an evaluation search unit 13 that performs image evaluation and search, and a personal information storage unit 14 that stores personal information such as an ID, a password, and an address in advance. , A keyword generation unit 16 that generates a keyword corresponding to the subject image included in the image data, a resolution conversion unit 18 that converts the resolution of the image data, and a score storage unit that stores a score for each keyword (hereinafter referred to as a keyword score). 19, an image database 20 for storing image data of newly posted images, an information database 22 for storing a link table 80 (see FIG. 5), a map database 24 for storing map data, and a traffic network included in the map data A transportation expense database 26 in which transportation expenses corresponding to the above are recorded is connected.

  Here, the personal information stored in the personal information storage unit 14 is registered in advance by the user of the terminal 8 that can communicate with the image evaluation server 4. The keyword generation unit 16 includes known image classification software capable of recognizing a subject image included in an image and generating a keyword corresponding to the recognized subject image.

  Further, the keyword score stored in the score storage unit 19 is appropriately updated based on, for example, the keyword search ranking. As a specific example, every time a keyword search is performed, 10 points are added to the keyword score ranked first, 8 points are added to the keyword score ranked second, and the keyword score ranked third. Add 5 points.

  FIG. 3 is a block diagram illustrating a configuration of the terminal 8 that can communicate with the image evaluation server 4 according to the embodiment. The terminal 8 includes a CPU 30. The CPU 30 includes a communication unit 32 that communicates with the image evaluation server 4 via the network 6, a GPS 34 that measures a current position based on a signal received from a GPS satellite, an LCD A display unit 38, an image sensor 41 that captures subject light and generates an image signal, and a memory card slot 43 in which a memory card 42 that stores image data is mounted are connected. Depending on the type of the terminal 8, the image pickup device 41 and the GPS 34 may not be provided, but a configuration that substitutes for these functions may be provided. That is, a personal computer or the like that does not include the image pickup device 41 or the GPS 34 may include a storage device that stores image data instead, and may only be able to specify approximate position information via the communication base station.

  FIG. 4 is a diagram showing an image evaluation site screen (hereinafter referred to as a site screen) provided by the image evaluation server 4 according to the embodiment. The site screen 50 includes an image display field 52 that displays an evaluation target image that is an evaluation target, a thumbnail image display field 54 that displays thumbnail images that are candidates for evaluation, and a post button 56 that is operated when posting an image. , An emergency GPS button 58, a map display field 60 for displaying a map based on map data, a keyword input field 62 for inputting a keyword for searching for an image, and the evaluation score of a keyword corresponding to a subject image in the evaluation target image Are displayed, an evaluation score display field 64, a comment display field 66 for displaying a comment assigned to the evaluation target image, and a transportation cost display field 68.

  FIG. 5 is a diagram showing a configuration of the link table 80 stored in the information database 22 of the image evaluation server 4 according to the embodiment. As shown in FIG. 5, the link table 80 stores a record for recording information regarding image data acquired by the image evaluation server 4. Here, each record includes an image link destination such as a URL of a site where an image corresponding to the record can be browsed, a keyword corresponding to a subject image included in the image data, an evaluation score assigned to the keyword, and an image. The last access date of the image, the shooting date of the image, and the like are recorded.

  Next, processing for posting an image to the image evaluation site provided by the image evaluation server 4 according to the embodiment will be described with reference to the flowchart shown in FIG. The flowchart is described with the CPU 10 of the image evaluation server 4 as a main component.

  First, when a user who wants to post an image accesses the image evaluation site from the terminal 8 (digital camera, smartphone, tablet terminal, mobile phone, personal computer), the site screen 50 shown in FIG. Is displayed. Next, when the user clicks the posting button 56, a posting screen (not shown) is displayed on the LCD display unit 38.

  Next, the user selects one or a plurality of image data from the image data stored in the memory card 42 of the terminal 8, and an image based on the selected image data is posted to the image evaluation site. . Here, to the image data stored in the memory card 42, a file name, shooting date information, shooting point position information, and the like are added as tag information.

  Next, the CPU 10 of the image evaluation server 4 uses the keyword generation unit 16 to generate a keyword corresponding to the subject image included in the image data of the posted image (step S1). For example, when an image A in which cherry blossoms and Mt. Fuji are reflected is uploaded, the keyword generating unit 16 recognizes that the image data of the image A includes cherry and Mt. Fuji subject images. The keyword “Mt. Fuji” is generated.

  Next, the CPU 10 stores the image data in the image database 20 (step S2). When the image data is stored in the image database 20, URL where the image A can be browsed, location information such as a storage destination directory name in the image database 20, and the like are determined.

  Next, CPU10 determines the evaluation score for every keyword produced | generated in step S2 by the evaluation search part 13 (step S3). First, the evaluation search unit 13 reads out the keyword score (for example, 5 points) corresponding to “Sakura” and the keyword score (for example, 4 points) corresponding to “Mt. Fuji” from the score storage unit 19. Then, for example, 2 points are added to each keyword score as a posting weight score to be added when an image is newly posted. In this case, the evaluation score of “Sakura” is determined to be 7 points (keyword score 5 points + posting weight score 2 points), and the evaluation score of “Mount Fuji” is 6 points (keyword score 4 points + posting weight score 2 points). .

  Next, the CPU 10 creates a record relating to this image in the link table 80 (step S4). For example, as shown in FIG. 5, a record “record 1” is created in the link table 80. Here, in “Record 1”, “image link destination” where the image can be browsed is recorded as “www.xxx.xxx”. Also, “Cherry Blossom” and “7 points” are recorded as “Keyword 1, Evaluation Score”, and “Mt. Fuji” and “6 Points” are recorded as “Keyword 2, Evaluation Score”. In addition, as the “last access date”, the date on which the image is posted is recorded as “2013/06/06”. Furthermore, the shooting date added to the image data is recorded as “2013/06/05” as “shooting date”, and the shooting point added to the image data as “shooting point”. Position information is recorded.

  Next, image evaluation processing at the image evaluation site provided by the image evaluation server 4 according to the embodiment will be described with reference to the flowchart shown in FIG. The flowchart is described with the CPU 10 of the image evaluation server 4 as a main component.

  First, on the site screen 50 (see FIG. 4), when a user who wants to evaluate an image inputs a keyword in the keyword input field 62, the CPU 10 of the image evaluation server 4 uses the evaluation search unit 13 to set the input keyword as a parameter. As a result, an image search in the image evaluation site is performed (step S21). For example, as shown in FIG. 4, when the keywords “sakura” and “Mt. Fuji” are entered in the keyword input field 62, the evaluation search unit 13 records the keywords “sakura” and “Mt. Fuji” from the link table 80. Extracted “Record 1” is extracted.

  The evaluation search unit 13 searches for images on the external search site 9 (search site A, search site B,...) In parallel with the search on the image evaluation site (step S22). For example, the evaluation search unit 13 transmits the keyword “sakura” and “Mt. Fuji” to the external search site 9 via the network 6 and a search request for executing a search using these keywords. When receiving the keyword information and the search request, the external search site 9 searches for an image using the received keyword information as a parameter.

  Here, a predetermined ranking is associated with the image data of the image detected as a result of the search. The external search site 9 transmits image data of images having a ranking within, for example, the top 30, to the image evaluation server 4 via the network 6. As to how many image data the external search site 9 transmits within, the predetermined instruction signal is transmitted from the image evaluation server 4 side when executing the search.

  Further, the predetermined ranking is calculated based on, for example, the hit frequency of the image when the image is searched during a specific period. Specifically, it is assumed that, for example, keyword search is performed 100 times in a specific period, image a is hit 10 times, image b is 8 times, and image c is hit 5 times. In this case, 10 points are given to the image a, 8 points are given to the image b, and 5 points are given to the image c. The ranking of images is as follows: image a is first, image b is second, image c is third, and so on.

  Next, the CPU 10 of the image evaluation server 4 receives the image data via the network 6 (step S23), and the keyword generating unit 16 generates a keyword corresponding to the subject image included in the image data (step S24). . Here, the keyword is generated in the same manner as in step S1 in which the image is posted. When a subject image other than “sakura” and “Mt. Fuji”, which are keywords at the time of search, is recognized from the image data, keywords corresponding to other subject images are further generated.

  Next, CPU10 determines the evaluation score for every keyword by the evaluation search part 13 similarly to the case of step S3. (Step S25). When the other subject is “automobile”, for example, the evaluation search unit 13 acquires the keyword score (for example, 3 points) corresponding to “automobile” from the score storage unit 19. Then, five points obtained by adding two posting weight points to the keyword points are determined as the “automobile” evaluation points.

  Next, the CPU 10 creates a record relating to the image data received in step S23 and adds it to the link table 80 (step S26). For example, as shown in FIG. 5, records “record 2” to “record n” are added to the link table 80. Here, each record is recorded in the same manner as at the time of posting (see step S4). That is, “Cherry, 7 points” (keyword score 5 points + posting weight score 2 points) is recorded in the “keyword 1, evaluation score 1” column of “record 1” to “record n”. In the “Evaluation score 2” column, “Mt. Fuji, 6 points” (keyword score 4 points + posting weight score 2 points) is recorded. If the image data corresponding to “record 2” includes a subject image of a car, “car, 5 points” (keyword score: 3 points) in the “keyword 3, evaluation score 3” column of “record 2” + Posting weight score 2 points) is recorded. In addition, “image link destination”, “last access date”, “shooting date”, and “shooting point” are recorded in “record 1” to “record n”, respectively.

  Next, the CPU 10 accesses the “image link destination” recorded in the “record 1” extracted in step S21 via the network 6, and the image corresponding to “record 1” from the “image link destination”. Get the data. Note that the image data corresponding to “record 2” to “record n” has already been acquired in step S23.

  Next, the CPU 10 recognizes the type of the terminal 8 used by the user (step S27). For example, when a keyword is input in step S21, the CPU 10 obtains information on the type of the keyword input source terminal 8, and based on this information, the type of the keyword input source terminal 8 (digital camera). , Smartphones, tablet devices, mobile phones, computers).

  Next, the CPU 10 refers to a terminal data table (not shown) that stores the type of the terminal 8 and the resolution of the image data in association with each other, and converts the image data to a resolution suitable for the size of the LCD display unit 38 of the terminal 8. To do. For example, when the terminal 8 is any one of a digital camera, a smartphone, a tablet terminal, and a mobile phone, the image data is converted so that the resolution is lower than when the terminal 8 is a personal computer. If the terminal 8 is a personal computer, the resolution of the image data need not be converted.

  Next, the CPU 10 generates thumbnail image data based on the image data, and displays the thumbnail image in the thumbnail image display field 54 of the terminal 8 (step S28). For example, as shown in FIG. 4, the thumbnail image A ′ of the image corresponding to “record 1”, the thumbnail image B ′ of the image corresponding to “record 2”,..., The thumbnail image of the image corresponding to “record n” N ′ is displayed in the thumbnail image display field 54.

  Next, a thumbnail image that the user wants to browse is selected from the thumbnail images displayed in the thumbnail image display field 54. For example, when viewing the thumbnail image A ′, the user operates the mouse while looking at the thumbnail image display field 54, moves the cursor to a position on the thumbnail image A ′, and clicks the thumbnail image A ′.

  When the thumbnail image A ′ is clicked, the CPU 10 identifies the image A, which is the original image of the thumbnail image A ′, as an evaluation target image, and displays the image A in the image display field 52 (step S29).

  Next, the CPU 10 detects the position information of the shooting point from “record 1”, “record 2” to “record n”, and reads out map data in a predetermined range including the position information from the map database 24. . Then, a map based on the map data is displayed on the map display field 60 (step S30), and a shooting location based on the position information is displayed in an overlapping manner at a corresponding position on the map with, for example, an X mark. Thereby, the user can easily recognize the shooting point of the image searched with the keyword “sakura, Mt. Fuji”. Next, the CPU 10 causes a mark 69 indicating that the image A is displayed in the image display field 52 to be displayed as a circle at the shooting location of the image A on the map.

  Next, CPU10 determines the evaluation score added to the image A by the evaluation search part 13 (step S31). Here, since the image A includes subject images of “sakura” and “mountain”, the CPU 10 corresponds to the keyword score of 5 points corresponding to “sakura” and “mountain” from the score storage unit 19. Read 4 keyword scores. It is assumed that the keyword score is not updated from the score when the image A is posted. Further, the subject images of “Sakura” and “Mt. Fuji” included in the image A may be recognized again by using the keyword generation unit 16 and corresponding keywords may be generated. Further, by referring to “Record 1” corresponding to the image A, keywords corresponding to the subject images of “Sakura” and “Mt. Fuji” included in the image A may be recognized.

  Next, the CPU 10 adds, for example, one point to each keyword score as a browsing weight score to be added when the image is browsed. The browsing weight score is set lower than the posting weight score. As a result, the evaluation score of “Sakura” is determined to be 6 points (5 keyword points + 1 browsing weight score), and the evaluation score of “Mt. Fuji” is determined to be 5 points (4 keyword scores + 1 browsing weight score). .

  Next, the CPU 10 adds the new evaluation score determined in step S31 to the evaluation score already recorded in “record 1” by the evaluation search unit 13 (step S32). That is, a new evaluation score of 6 is added to the “evaluation score of 1” shown in FIG. 5, and the current “7 points” is updated to “13 points”. Similarly, a new evaluation score of 5 points is added to “evaluation score of 2”, and the current “6 points” is updated to “11 points”. Further, the CPU 10 updates the “last access date” of “record 1”.

  Next, the CPU 10 displays the updated evaluation score for each keyword in the evaluation score display column 64 (step S33). For example, as shown in FIG. 4, “Sakura evaluation score 13 points” and “Mt. Fuji evaluation score 11 points” are displayed. Thus, the user can easily recognize the evaluation score for each subject image shown in the current image A.

  According to the image evaluation system 2 according to this embodiment, a keyword corresponding to a subject image included in image data is generated, and an evaluation score is determined for each generated keyword. Can do. In addition, when an image is searched on the image evaluation site provided by the image evaluation server 4, an image search on the external search site 9 is also executed in parallel. Can be evaluated.

  Further, since the newly posted image and the image acquired from the external search site 9 are evaluated using the same algorithm, objective image evaluation is possible. In addition, since the record of the image with the highest ranking is recorded in the link table 80 in the external search site 9, a record of a useful image having a high popularity is stored in the link table 80. In addition, every time an image is searched, the image records acquired from the external search site 9 are sequentially stored in the link table 80, so that a huge number of records are efficiently recorded in a short time. The link table 80 can be completed. When the link table 80 is completed, there is no need to perform an image search at the external search site 9 in parallel when an image is searched at the image evaluation site, and the image search and browsing is completed only at the image evaluation site. Can be made.

  In addition, since the keyword score is appropriately updated based on the search ranking, it is possible to perform an evaluation reflecting the popularity of the current image. In addition, by setting the posting weight score higher than the browsing weight score, it is possible to prevent the evaluation score of an image newly posted and not browsed from being unduly lowered. Further, since the resolution of the image data is adjusted according to the size of the display unit of the terminal 8, the data reception amount on the terminal 8 side can be reduced, and the image display processing speed can be increased.

  In the above-described embodiment, when the “last access date” recorded in the record in the link table 80 is not updated for a predetermined time or more, the evaluation score recorded in the record may be deducted by a predetermined score. . For example, if the “Last Access Date” of “Record 1” shown in FIG. 5 is not updated for one year, the evaluation points of “Sakura, 7 points” and “Mt. Fuji, 6 points” will be decremented by 5 points respectively. “2 points” and “Mt. Fuji, 1 point”. If the total evaluation score of each keyword recorded in “record 1” (here, 3 points (2 points + 1 point)) falls below a predetermined evaluation score (for example, 4 points), the record is stored in the link table. You may delete from 80. Thereby, the evaluation is not performed, and the record of the unpopular image is periodically deleted. Whether or not the “last access date” has been updated is determined at a pace of once every several months, for example.

  In the above-described embodiment, when an image search is performed at the external search site 9, information related to the image score is received together with image data of the searched image, and image evaluation is performed using the received information related to the score. May be performed.

  For example, assume that the user performs a search using the keyword “Tokyo Tower”. In this case, the evaluation search unit 13 first searches the image evaluation server 4 for a record in which the keyword “Tokyo Tower” is recorded from the link table 80. As the evaluation score of Tokyo Tower, for example, record X (not shown) in which “25 points” is recorded, record Y (not shown) in which “10 points” are recorded, and “8 points” are recorded. Records Z (not shown),... Are extracted in descending order of evaluation score. Here, it is assumed that evaluation scores other than Tokyo Tower are not recorded in the records X to Z.

  Next, the evaluation search unit 13 searches the external search site 9 for an image using the keyword “Tokyo Tower”. Then, image data of an image having a high score ranking (for example, the top 30 or more) is acquired from the searched images via the network 6. For example, the image data of the image α with the score “100 points”, the image data of the image β with the score “80 points”, the image data of the image γ with the score “60 points”,... Obtain image data along with information about the score.

  Next, the evaluation search unit 13 adjusts the score acquired via the network 6 in accordance with the evaluation score recorded in the link table 80. For example, the score “100 points” of image α (score ranking first) is matched with “25 points” recorded in record X (first ranking evaluation score), and the score “100 points” of image α is set to 1. It is changed to “25 points” which is a score of / 4. In accordance with this, the score of ranking 2nd or lower is also changed to a score multiplied by 1/4. In this case, the score of the image β is changed to “20 points”, the score of the image γ is “12 points”,.

  Next, the CPU 10 of the image evaluation server 4 compares the evaluation score recorded in the records X to Z,... With the score after the change of the images α to γ,. If the score value is equal to or greater than the evaluation score value, an image record corresponding to the score is created and added to the link table 80.

  For example, since the score “25 points” (first ranking in the score) after the change of the image α is the same as “25 points” (first ranking in the evaluation score) recorded in the record X, the CPU 10 Create “Record α” of α. Then, the posting weight score of 2 points is added to the score “25 points”, and “record α” of the image α in which “Tokyo Tower, 27 points” is recorded as “keyword, evaluation score” is added to the link table 80.

  Similarly, since the score “20 points” (second score ranking) after the change of the image β is equal to or higher than “10 points” (second ranking evaluation score) recorded in the record Y, the CPU 10 Create “Record β” of β. Then, 2 posting weight points are added to the score “20 points”, and “record β” of the image β in which “Tokyo Tower, 22 points” is recorded as “keyword, evaluation score” is added to the link table 80. Since the score “12 points” (score ranking 3rd) after the change of the image γ is lower than “8 points” (ranking ranking 3rd in the evaluation score) recorded in the record Z, the record of the image γ is Not created.

  Thus, when determining the evaluation score of the image searched on the external search site 9, by using the score calculated on the external search site 9, it is possible to generate keywords without using image data. Records corresponding to images searched on the external search site 9 can be easily created. In addition, as a score of an image, for example, when a search is performed using Google AdWords (registered trademark), the image evaluation server 4 may receive a quality score obtained as a search result. Moreover, when using a score, you may perform keyword generation by the keyword production | generation part 16 collectively.

  Moreover, in step S24 of the above-described embodiment, when the keyword input by the user is different from the keyword generated by the keyword generation unit 16, it is not necessary to create a record relating to the image data received in step S23. For example, it is assumed that the user performs a search using “sakura” and “Mt. Fuji” as keywords, and image data of images with the ranking within the top 30 is acquired from the external search site 9. Of these, from the image data K, only the keyword “automobile” is generated, and the keywords “sakura” and “Mt. Fuji” are not generated. In such a case, the CPU 10 may not create a record relating to the image data K.

  Moreover, in the above-mentioned embodiment, when a user submits an image, a comment of the image may be input. In this case, the CPU 10 records the comment itself in a record corresponding to the posted image. Then, when this image is browsed and displayed in the image display field 52, a comment is displayed in the comment display field 66 as shown in FIG. When creating a record, the evaluation retrieval unit 13 reads the keyword score corresponding to the keyword included in the comment from the score storage unit 19 and records the score obtained by adding the posted weight score to the keyword score as the evaluation score. May be. Thus, by reading out a keyword from a comment, it is possible to accurately extract a keyword corresponding to the subject intended by the user.

  In the above-described embodiment, information related to transportation expenses may be displayed in the transportation expense display column 68. For example, when an image is displayed in the image display field 52, the CPU 10 acquires current position information of the terminal 8 by the GPS 34. Next, the CPU 10 searches for a route from the current position of the terminal 8 to the shooting point by referring to the traffic network included in the map data. Next, the CPU 10 refers to the transportation cost database 26 to calculate a transportation cost from the current position to the shooting point when each searched route is used. As shown in FIG. 4, the transportation cost for each route is calculated. Is displayed in the transportation cost display column 68.

  Next, when a predetermined route and transportation cost are selected by the user, the CPU 10 displays details such as time and transfer information required when the route is selected in the vicinity of the transportation cost display column 68. Thereby, the user can easily know the route to the shooting point of the evaluated image and the transportation cost.

  In the above-described embodiment, as shown on the left side of FIG. 8, the size of the mark 69 displayed on the map may be changed according to the size of the evaluation score for the image. For example, as shown in FIG. 4, if the total of the evaluation scores displayed in the evaluation score display field 64 is 24 points (13 points + 11 points), the evaluation score displayed in the evaluation score display field 64 The mark 69 displayed in the map display column 60 is made larger than when the total is 13 points (7 points + 6 points).

  Further, as shown on the right side of FIG. 8, the color of the route displayed in the transportation expense display column 68 may be changed according to the amount of transportation expense. For example, as shown in FIG. 4, three routes of “route 1: transportation cost 1500 yen”, “route 2: transportation cost 1000 yen”, and “path 3: transportation cost 900 yen” are displayed in the transportation cost display field 68. Suppose that it was done. In this case, “route 1: transportation cost 1500 yen” is displayed in red, “path 2: transportation cost 1000 yen” is displayed in yellow, and “path 3: transportation cost 900 yen” is displayed in blue. Thereby, the user can easily determine whether it is worth going to the shooting point of the evaluated image.

  In the above-described embodiment, when a thumbnail image is continuously clicked within a predetermined time (for example, when double-clicked), the image may not be evaluated. Thereby, it is possible to prevent an image from being erroneously evaluated due to an operation mistake or the like.

  In the above embodiment, when the emergency GPS button 58 shown in FIG. 4 is operated, the terminal 8 acquires the current position information of the terminal 8 by the GPS 34, and displays a map or aerial photograph in a predetermined range including the current position. You may make it display on the image display column 52. FIG. And you may display the mark which shows the present self position in the corresponding position on a map.

  Further, in the above-described embodiment, when a map is displayed in the map display field 60, the display range of the map may be adjusted according to the degree of congestion of the shooting spots displayed on the map. Here, the degree of crowding of the shooting points is determined as follows, for example. First, a difference value between the shooting point of the image displayed in the image display field 52 and the other shooting point displayed on the map is calculated. Next, a variance value is calculated based on the difference value, and the degree of congestion is determined according to the size of the variance value. As a result, as shown in FIG. 9 (a), when the shooting points displayed on the map are dense, as shown in FIG. 9 (b), the display range of the map is corrected and photographed with an appropriate degree of congestion. The point can be displayed on the map.

  Further, in the above-described embodiment, the user may select the shooting point displayed in the map display field 60 so that the image at the selected shooting point is displayed in the image display field 52. For example, it is assumed that the image A is displayed in the image display field 52 and the mark 69 is displayed at the shooting point of the image A. Here, as shown in FIG. 10A, when the user moves the mark 69 to the shooting point of the image B in the map display field 60, the image is displayed in the image display field 52 as shown in FIG. B is displayed.

  Further, in the above-described embodiment, as shown in FIG. 4, it is assumed that the site screen 50 is displayed on a personal computer or the like having a large display surface. When a tablet terminal, a mobile phone or the like is used as the terminal 8, a site screen 100 as shown in FIG. 11 may be displayed. Here, in the same manner as the site screen 50, the site screen 100 includes a thumbnail image display field 54, an image display field 52 for displaying an image to be evaluated, a post button 56 operated when posting an image, a map A map display field 60 for displaying a map based on the data and a keyword input field 62 for inputting a keyword when searching for an image are included.

  In the above-described embodiment, when the image evaluation server 4 does not include the map data storage unit 29 and performs image evaluation processing, the CPU 20 acquires map data from a predetermined map site or the like via the network 6. Also good. Similarly, the transportation cost corresponding to the transportation network included in the map data may be acquired via the network 6 when the map data is acquired.

  In the above-described embodiment, the image classification software built in the keyword generation unit 16 and other predetermined software built in the evaluation search unit 13 and the resolution conversion unit 18 are upgraded as appropriate.

DESCRIPTION OF SYMBOLS 2 ... Image evaluation system, 4 ... Image evaluation server, 6 ... Network, 8 ... Terminal, 9 ... External search site, 10 ... CPU, 12 ... Communication part, 14 ... Personal information storage part, 16 ... Keyword generation part, 18 ... Resolution conversion unit, 19 ... score storage unit, 20 image database, 22 information database, map database 24, transportation cost database 26, 30 ... CPU, 34 ... GPS, 38 ... LCD display unit, 41 ... image sensor, 42 ... memory card

Claims (12)

An image acquisition unit for acquiring image data;
A keyword acquisition unit for acquiring a keyword corresponding to at least one subject image included in the image data;
A determination unit that determines an evaluation score when the image data is acquired by the image acquisition unit for each of the keywords acquired by the keyword acquisition unit;
An image evaluation server comprising: a recording unit that records the evaluation score for each of the keywords determined by the determination unit in association with the image data. A specifying unit for specifying an evaluation target image from images based on the image data,
The keyword acquisition unit acquires the keyword corresponding to at least one of the subject images included in the image data of the evaluation target image,
The determining unit determines an evaluation score when the evaluation target image is specified by the specifying unit for each keyword acquired by the keyword acquiring unit,
The image evaluation server according to claim 1, wherein the recording unit adds the evaluation score for each keyword determined by the determination unit to the evaluation score already recorded in the recording unit. A search unit for searching for an image using the input keyword,
A score storage unit for storing a keyword score for each keyword referred to when the evaluation score for each keyword is determined by the determination unit;
The image evaluation server according to claim 1, further comprising: a changing unit that changes the keyword score stored in the score storage unit based on a search result by the search unit. A search instruction unit for instructing an image search using the input keyword to an external image search site when a search is performed by the search unit;
The image evaluation server according to claim 3, wherein the image acquisition unit acquires the image data of an image searched at the external image search site. The image acquisition unit acquires a score associated with an image searched on the external image search site,
The image evaluation server according to claim 4, wherein the determination unit determines the evaluation score for each of the keywords using the score.   The determination unit determines the evaluation score when the image data is acquired by the image acquisition unit to be higher than the evaluation score when the evaluation target image is specified by the specifying unit. The image evaluation server as described in any one of Claims 2-5. A date and time recording unit that causes the recording unit to record at least one of the date and time when the image data was acquired and the date and time when the evaluation target image was specified;
When a predetermined time has elapsed from the date and time recorded in the recording unit, a deduction unit for deducting a predetermined score from the evaluation score recorded in the recording unit;
And a deletion unit that deletes a record related to the image data associated with the evaluation score from the recording unit when the evaluation score is equal to or lower than a predetermined evaluation score as a result of the deduction by the deduction unit. The image evaluation server according to any one of claims 2 to 6, wherein the image evaluation server is characterized in that A type acquisition unit that acquires information about the type of the terminal from a terminal used when a user browses an image based on the image data;
An adjustment unit that adjusts the resolution of the image data according to the type of the terminal acquired by the type acquisition unit;
The image evaluation server according to claim 1, wherein an image based on the image data adjusted to a predetermined resolution by the adjustment unit is displayed on a display unit of the terminal. A map acquisition unit that acquires map data
In the recording unit, shooting position information of an image based on the image data is recorded,
The image evaluation according to claim 1, wherein the shooting position is displayed on the map in a display unit of a terminal used when a user browses an image based on the image data. server.   The image evaluation server according to claim 9, further comprising a scale changing unit that changes a scale of a map displayed on the display unit in accordance with a degree of density of the photographing positions. On the map, a mark indicating the shooting position of the evaluation target image is displayed,
The image evaluation server according to claim 9 or 10, wherein a size of the mark is changed according to a size of the evaluation score. A map acquisition unit that acquires map data
The display unit of a terminal used when a user browses an image based on the image data displays a position of the terminal measured by GPS on the map. The image evaluation server according to item.

Images