Estimation device, estimation method, and program

Abstract

The present invention provides an estimation device (10) comprising: an acquisition unit (11) that acquires a posted image to be processed that was posted on the Internet; a related area identification unit (12) that identifies an area related to the target contributor who posted the posted image to be processed; a selection unit (13) that selects a reference image to be matched with the posted image to be processed, on the basis of the result of identifying the area related to the target contributor; and a location estimation unit (14) that estimates a location indicated by the posted image to be processed, on the basis of the result of matching the selected reference image with the posted image to be processed.

Description

Technical Field

[0001] The present invention relates to an estimation device, an estimation method, and a program.

Background Art

[0002] Patent Document 1 discloses a method of associating an image with a facility. In this method, when a query image is received, a list of candidate facilities is identified from social media using GPS (Global Positioning System). Then, after a business-related concept is detected from the image content, the business-related concept is collated with those extracted from reviews and images stored in a database. And the facility related to the image and review having the highest consistency is returned as the most likely facility.

[0003] Also, Patent Documents 2 and Non-Patent Documents 1 to 4 disclose techniques for estimating a user's activity range.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Non-Patent Documents

[0005]

Non-Patent Document 1

Non-Patent Document 2

[0006] There is a need for a technique to identify the location indicated by a posted image on the Internet (the location of the subject in the posted image). When metadata (location information) is not associated with the posted image, for example, means such as collating the posted image with a reference image with location information and identifying the location indicated by the posted image based on the collation result are known. However, in this case, it is basically necessary to collate the posted image with a huge number of reference images prepared to cover all locations in the world. Therefore, the processing time increases.

[0007] The present invention aims to shorten the time required for processing in a means for identifying the position indicated by a posted image based on the collation result between a reference image and the posted image.

Means for Solving the Problem

[0008] According to the present invention, an acquisition means for acquiring a posted image to be processed posted on the Internet, a related area specifying means for specifying a related area of a target poster who posted the posted image to be processed, a selection means for selecting a reference image to be collated with the posted image to be processed based on the specification result of the related area of the target poster, a position estimation means for estimating the position indicated by the posted image to be processed based on the collation result between the selected reference image and the posted image to be processed, a estimation device having the above is provided.

[0009] Also, according to the present invention, a computer performs an acquisition step of acquiring a posted image to be processed posted on the Internet, a related area specifying step of specifying a related area of a target poster who posted the posted image to be processed, a selection step of selecting a reference image to be collated with the posted image to be processed based on the specification result of the related area of the target poster, a position estimation step of estimating the position indicated by the posted image to be processed based on the collation result between the selected reference image and the posted image to be processed, a estimation method for executing the above is provided.

[0010] Also, according to the present invention, a computer is equipped with an acquisition means for acquiring a posted image to be processed posted on the Internet, a related area specifying means for specifying a related area of a target poster who posted the posted image to be processed, Selection means for selecting a reference image to be compared with the input image to be processed based on the identification result of the relevant area of the target submitter, and Position estimation means for estimating the position indicated by the input image to be processed based on the comparison result between the selected reference image and the input image to be processed, A program that functions as such is provided.

Advantages of the Invention

[0011] According to the present invention, in the means for identifying the position indicated by the input image based on the comparison result between the reference image and the input image, the time required for processing can be shortened.

Brief Description of the Drawings

[0012]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Embodiments for Carrying Out the Invention

[0013] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In all the drawings, the same components are denoted by the same reference numerals, and the description will be omitted as appropriate.

[0014] <Summary> The estimation device of this embodiment identifies the position indicated by the posted image (still image and / or video) to be processed based on the collation result between the reference image and the posted image to be processed. And the estimation device has features for shortening the time required for such processing. Specifically, the estimation device identifies an area related to the target poster who posted the posted image to be processed (hereinafter, "related area"), and sets the reference image selected based on the related area as the target to be collated with the posted image to be processed. By narrowing down the reference images to be collated in this way, the time required for processing is shortened.

[0015] Note that the related area of the target poster is identified based on at least one of the account information of the target poster, the posted image to be processed, the past posted images of the target poster, the position information associated with the past posted images of the target poster, the posted text posted together with the posted image to be processed, the past posted texts of the target poster, the language used by the target poster, the posting timing of the target poster, the activity area of the target poster, and the related areas of related posters having a predetermined relationship with the target poster. By identifying the related area of the target poster based on such information, it is possible to accurately identify the area truly related to the target poster with high precision.

[0016] <Configuration> Next, the configuration of the estimation device 10 will be described.

[0017] 「Overall Configuration」 As shown in FIG. 1, the estimation device 10 is configured to be communicable with the social media system 20. The estimation device 10 receives various information from the social media system 20 and executes various processes based on the received information.

[0018] The social media system 20 is a system that provides social media services such as SNS (Social Networking Service). The social media system 20 may include one social media service or multiple social media services. A social media service is an online service that can transmit (publicize) information and communicate among multiple accounts (users) on the Internet (online). The social media service is not limited to SNS and includes messaging services such as chat, blogs and bulletin boards (forum sites), video sharing sites and information sharing sites, social games and social bookmarks, etc.

[0019] For example, the social media system 20 includes a server on the cloud and user terminals. The server may be a social media server or a web server. The user terminal logs in with the user's account via the API (Application Programming Interface) provided by the server, inputs and browses posts, and also registers the connections of accounts such as friendship and follow relationships.

[0020] "Functional Configuration" Next, the functional configuration of the estimation device 10 will be described. As shown in FIG. 2, the estimation device 10 includes an acquisition unit 11, a related area identification unit 12, a selection unit 13, a position estimation unit 14, and a storage unit 15. Note that, as shown in FIG. 3, the estimation device 10 may not have the storage unit 15. In this case, an external device configured to be communicable with the estimation device 10 includes the storage unit 15.

[0021] The acquisition unit 11 acquires a target posted image to be processed, which is one of the posted images posted on the Internet, from the social media system 20. For example, the acquisition unit 11 acquires, as the target posted image to be processed, one posted image specified / selected / determined, etc. by an input from an operator who operates the estimation device 10. The operator browses a plurality of posted images and specifies / selects / determines, etc. the posted image for which the operator wants to know the position shown in the image, as the target posted image to be processed. For example, the operator specifies / selects / determines, etc. the posted image in which a predetermined event (incident, accident, disaster, etc.) or a subject is captured, as the target posted image to be processed.

[0022] The related area specifying unit 12 specifies the related area of the target poster who posted the target posted image to be processed. The related area of the target poster is an area related to the target poster. The related area of the target poster is specified based on at least one of the account information of the target poster, the target posted image to be processed, the past posted images of the target poster, the position information associated with the past posted images of the target poster, the posted text posted together with the target posted image to be processed, the past posted texts of the target poster, the language used by the target poster, the posting timing of the target poster, the activity area of the target poster, and the related areas of related posters having a predetermined relationship with the target poster. This will be described in detail below.

[0023] -Specification based on the account information of the target poster- The related area specifying unit 12 acquires the account information of the target poster from the social media system 20. The account information includes, for example, location information of the place of residence (place name, address, postal code, telephone number, etc.), location information of the place of origin (place name, address, postal code, telephone number, etc.), location information of the workplace or school (place name, address, postal code, telephone number, etc.), location information of the alma mater (place name, address, postal code, telephone number, etc.), and the like. The related area specifying unit 12 specifies, as the related area of the target poster, the area indicated by such account information or an area further including the periphery thereof. The account information may include, in addition to the above information, other information described in the profile. Note that the related area specifying unit 12 may acquire only the account information that is set to be publicly available to other users, or may acquire both the account information that is set to be publicly available to other users and the account information that is not set to be publicly available to other users.

[0024] There are various algorithms for specifying the related area from the account information, and any method can be adopted. For example, the area indicated by the place name may be specified as the related area of the target poster, or the area indicated by the place name and the area within a predetermined distance therefrom may be specified as the related area of the target poster. Also, the area within a predetermined distance from the point indicated by the address may be specified as the related area of the target poster. Also, the area indicated by the postal code may be specified as the related area of the target poster, or the area indicated by the postal code and the area within a predetermined distance therefrom may be specified as the related area of the target poster. Also, the area indicated by the telephone number (for example, the area indicated by the area code outside the city, etc.) may be specified as the related area of the target poster, or the area indicated by the telephone number and the area within a predetermined distance therefrom may be specified as the related area of the target poster.

[0025] -Specification based on the posting image to be processed and the past posting images of the target poster (1)- The related area specifying unit 12 acquires at least one of the input posting image and the past posting images of the target poster from the social media system 20. Then, the related area specifying unit 12 specifies the related area of the target poster based on at least one of the subject of the input posting image and the subjects of the past posting images of the target poster.

[0026] The related area specifying unit 12 specifies a rough area indicated by the posting image based on the subject of the posting image. Then, the related area specifying unit 12 specifies this specified rough area as the related area of the target poster.

[0027] The units of the rough areas are various. For example, it may be a country. Also, the rough area may be a unit smaller than a country. For example, in the case of Japan, prefectures, municipalities, western Japan, eastern Japan, Tohoku, Kanto, Chubu, Kinki, Chugoku, Shikoku, Kyushu, etc. are exemplified.

[0028] For example, the related area specifying unit 12 can specify the rough area indicated by the posting image based on the architectural style of the building that is the subject, the presence or absence of a telegraph pole that is the subject, the traffic direction of the vehicle that is the subject, the type of plant that is the subject, the type of product that is the subject, the clothes of the person that is the subject, the information on the license plate of the vehicle that is the subject, the characters included in the posting image, the place name or address indicated by the characters, the language of the characters, symbols, numbers, etc.

[0029] For example, a means for specifying the rough area indicated by the posting image based on an estimation model created by machine learning can be considered, but it is not limited to this.

[0030] In addition, as means for specifying the rough area indicated by the posting image based on the characters detected from the posting image, for example, the following processing can also be considered.

[0031] The related area specifying unit 12 specifies the rough area indicated by the posting image based on a predetermined keyword included in the characters detected from the posting image.

[0032] Keywords can be used to narrow down the area. For example, but not limited to, place names, facility names, product names, famous people's names, company names, team names, event names, etc. For example, the relevant area identification unit 12 may identify the area indicated by the place name as the relevant area of the target submitter, or may identify the area indicated by the place name and the area within a predetermined distance therefrom as the relevant area of the target submitter. In addition, the relevant area identification unit 12 may identify the area where the facility, company, or team indicated by the keyword exists or operates as the relevant area of the target submitter. In addition, the relevant area identification unit 12 may identify the area where the event indicated by the keyword is held as the relevant area of the target submitter. In addition, the relevant area identification unit 12 may identify the area where the product indicated by the keyword is sold as the relevant area of the target submitter. In addition, the relevant area identification unit 12 may identify the area where the famous person indicated by the keyword is active as the relevant area of the target submitter.

[0033] The relevant area identification unit 12 may identify the area indicated by the keyword included in the characters detected from the submitted image based on a database that pre-associates keywords with location information (information indicating areas such as areas indicated by place names, areas where facilities, companies, or teams exist or operate, areas where events are held, areas where products are sold, areas where famous people are active, etc.). Alternatively, the relevant area identification unit 12 may perform a predetermined search such as a web search using the keyword included in the characters detected from the submitted image as a key, and acquire location information such as that stored in the above-described database from the search results.

[0034] When there are multiple keywords, the area included in at least one of the areas identified based on each of the multiple keywords may be identified as the relevant area of the target submitter.

[0035] In addition, when there are multiple submitted images, the area included in at least one of the approximate areas identified based on each of the multiple submitted images may be identified as the relevant area of the target submitter.

[0036] -Specification based on the posting image to be processed and the past posting images of the target poster (2)- The related area specifying unit 12 acquires at least one of the posting image (moving image) to be processed and the past posting images (moving images) of the target poster from the social media system 20. Then, the related area specifying unit 12 specifies the related area of the target poster based on at least one of the voice associated with the posting image to be processed and the voice associated with the past posting images of the target poster.

[0037] The related area specifying unit 12 specifies the approximate area indicated by the posting image based on the voice associated with the posting image. Then, the related area specifying unit 12 specifies this specified approximate area as the related area of the target poster. The concept of the approximate area is as described above.

[0038] For example, the related area specifying unit 12 can specify the approximate area indicated by the posting image based on the ambient sound (signal machine such as the sound emitted from installations such as level crossings), the language used, the music being played, etc. included in the voice associated with the posting image.

[0039] For example, means for specifying the approximate area indicated by the posting image based on an estimation model created by machine learning can be considered, but it is not limited to this.

[0040] When there are a plurality of posting images (moving images), an area included in at least one of the approximate areas specified based on each of the plurality of posting images may be specified as the related area of the target poster.

[0041] -Specification based on the location information associated with the past posting images of the target poster- The related area specifying unit 12 acquires the location information associated with the past posting images of the target poster from the social media system 20. Then, the related area specifying unit 12 specifies the related area of the target poster based on the location information associated with the past posting images of the target poster.

[0042] The position information associated with the submitted image is metadata called, for example, a geotag, and indicates the position in terms of latitude and longitude. The related area specifying unit 12 may specify, as the related area of the target submitter, an area within a predetermined distance from the position indicated by the position information associated with the past submitted images of the target submitter. Also, when there are a plurality of past submitted images associated with position information, an area included in at least one of the related areas specified based on each of the plurality of past submitted images may be specified as the related area of the target submitter.

[0043] -Specification based on the submission text submitted together with the submission image to be processed and the past submission texts of the target submitter The related area specifying unit 12 acquires at least one of the submission text submitted together with the submission image to be processed and the past submission texts of the target submitter from the social media system 20. Then, the related area specifying unit 12 specifies the related area of the target submitter based on at least one of the keywords included in the submission text submitted together with the submission image to be processed and the keywords included in the past submission texts of the target submitter.

[0044] Keywords can be used for narrowing down the area, and examples include, but are not limited to, place names, facility names, product names, famous people's names, company names, team names, event names, etc. For example, the related area specifying unit 12 may specify, as the related area of the target submitter, an area indicated by a place name, or may specify, as the related area of the target submitter, an area indicated by a place name and an area within a predetermined distance therefrom. Also, the related area specifying unit 12 may specify, as the related area of the target submitter, an area where a facility, company, or team indicated by a keyword exists or conducts activities. Also, the related area specifying unit 12 may specify, as the related area of the target submitter, an area where an event indicated by a keyword is held. Also, the related area specifying unit 12 may specify, as the related area of the target submitter, an area where a product indicated by a keyword is sold. Also, the related area specifying unit 12 may specify, as the related area of the target submitter, an area where a famous person indicated by a keyword is active.

[0045] The related area identification unit 12 may identify the area indicated by the keyword extracted from the posted text based on a database in which keywords are associated in advance with location information (information indicating an area indicated by a place name, an area where a facility, company, or team exists or conducts activities, an area where an event is held, an area where a product is sold, an area where a famous person conducts activities, etc.). Alternatively, the related area identification unit 12 may perform a predetermined search such as a web search using the keyword extracted from the posted text as a key, and acquire location information such as that stored in the above-described database from the search results.

[0046] When there are multiple keywords, an area included in at least one of the areas identified based on each of the multiple keywords may be identified as the related area of the target poster.

[0047] - Identification based on the language used by the target poster - The related area identification unit 12 acquires from the social media system 20 at least one of the account information of the target poster, the posted text posted together with the posted image to be processed, and the past posted texts of the target poster. Then, the related area identification unit 12 identifies the language used by the target poster in this information. Then, the related area identification unit 12 identifies the related area of the target poster based on the identified language. Specifically, the related area identification unit 12 identifies the area where the identified language is used as the related area of the target poster.

[0048] The related area identification unit 12 can identify the area where the identified language is used based on a database in which the language used is associated in advance with location information (information indicating the area where each language is used).

[0049] - Identification based on the posting timing of the target poster - The related area identification unit 12 acquires information indicating the past posting timing of the target poster from the social media system 20. Then, the related area identification unit 12 identifies the related area of the target poster based on the posting timing of the target poster.

[0050] For example, it is considered that the timing when there is little or no contribution is late at night, and the timing when there is a contribution is during the day. Based on such regularity, the time zone of the area where the target contributor exists is specified, and the area corresponding to the specified time zone is specified as the related area of the target contributor.

[0051] -Specification based on the activity area of the target contributor- The related area specifying unit 12 acquires various types of information regarding the target contributor from the social media system 20. Then, based on the various types of information, the related area specifying unit 12 specifies the activity area of the target contributor. And the related area specifying unit 12 specifies the specified activity area of the target contributor as the related area of the target contributor.

[0052] The specification of the activity area of the target contributor can be realized by using any conventional technique. For example, the techniques disclosed in Patent Document 2 and Non-Patent Documents 1 to 4 may be used, or other techniques may be used. For example, based on the position distribution of the target contributor specified based on the account information of the target contributor and the position distribution specified based on the account information of the related contributor having a predetermined relationship with the target contributor the activity area of the target contributor may be specified based on the position distribution of the related contributor having a predetermined relationship with the target. The position distribution is generated by a non-parametric method such as a kernel density estimation function, for example.

[0053] -Specification based on the related area of the related contributor having a predetermined relationship with the target contributor- A related contributor is a person who is connected to the target contributor on the service provided by the social media system 20. How to define which people are connected to each other depends on the functions provided by the social media system 20. For example, a related contributor is at least one of a person who follows the target contributor, a person who is followed by the target contributor, a person who has a mutual follow relationship with the target contributor, a person who mentions (replies or mentions) the posted image or post text posted by the target contributor, and a person who is mentioned (replied or mentioned) by the target contributor in the posted image or post text.

[0054] Based on the information obtained from the social media system 20, the related area specifying unit 12 specifies related contributors who have a predetermined relationship with the target contributor. Note that the social media system 20 may specify related contributors who have a predetermined relationship with the target contributor and transmit information (such as a list) indicating the specified related contributors to the estimation device 10.

[0055] The related area specifying unit 12 specifies the related area of related contributors who have a predetermined relationship with the target contributor as the related area of the target contributor. The related area of a related contributor is specified by the same method as the method for specifying the related area of the target contributor described above. That is, the related area of a related contributor is 、 related specified based on the account information of the related contributor, the past posted images of the related contributor, the position information associated with the past posted images of the related contributor, the past posted texts of the related contributor, the language used by the related contributor, the posting timing of the related contributor, the activity area of the related contributor, and the like.

[0056] Note that the related area specifying unit 12 may specify the related area of the target contributor by combining two or more of the above-described multiple types of methods. In this case, an area included in at least one of the related areas of the target contributor specified by each method may be specified as the related area of the target contributor.

[0057] Returning to FIGS. 2 and 3, the selection unit 13 selects a reference image to be compared with the input image to be processed based on the specific result of the related area of the target poster. As shown in FIG. 4, in advance, each of a plurality of reference images is stored in the storage unit 15 in association with position information. The position information associated with the reference image indicates the position shown by each reference image. The position information may be indicated by latitude and longitude, or may be indicated by an address or the like.

[0058] The selection unit 13 selects a reference image associated with position information having a predetermined relationship with the related area of the target poster from among the plurality of reference images stored in the storage unit 15. For example, the selection unit 13 selects a reference image associated with position information indicating a point within the related area of the target poster.

[0059] Based on the comparison result between the reference image selected by the selection unit 13 and the input image to be processed, the position estimation unit 14 estimates the position shown by the input image to be processed. Specifically, the position estimation unit 14 compares each of the reference images selected by the selection unit 13 with the input image to be processed, and calculates a comparison score. Then, the position estimation unit 14 outputs the position information associated with the reference image whose comparison score satisfies a predetermined condition as information indicating the position of the input image to be processed.

[0060] The comparison score indicates the similarity between two images. The comparison and the calculation of the comparison score can be realized using any conventional technique.

[0061] When it is indicated that the two images are more similar as the comparison score is larger, the position estimation unit 14 outputs the position information associated with the reference image whose comparison score is equal to or greater than the threshold value as information indicating the position of the input image to be processed. The output of the information is realized via any output device such as a display, a speaker, a projection device, a printer, a communication device, or the like.

[0062] Next, an example of the processing flow of the estimation device 10 will be described with reference to the flowchart of FIG. 5.

[0063] First, the estimation device 10 acquires a target posting image to be processed, which is one of the posting images posted on the Internet, from the social media system 20 (S10). For example, the acquisition unit 11 acquires, as the target posting image to be processed, one posting image specified / selected / determined, etc. by an operator who operates the estimation device 10.

[0064] Next, the estimation device 10 specifies the related area of the target poster who posted the target posting image to be processed (S11). The estimation device 10 specifies the related area of the target poster based on at least one of the account information of the target poster, the target posting image to be processed, the past posting images of the target poster, the position information associated with the past posting images of the target poster, the posting text posted together with the target posting image to be processed, the past posting texts of the target poster, the language used by the target poster, the posting timing of the target poster, the activity area of the target poster, and the related areas of related posters having a predetermined relationship with the target poster.

[0065] Next, the estimation device 10 selects a reference image to be compared with the target posting image to be processed based on the specification result of the related area of the target poster (S12) 。 For example, the estimation device 10 selects a reference image associated with position information indicating a location within the related area of the target poster. The estimation device 10 selects one or more reference images associated with the position within the related area of the target poster specified in S11.

[0066] Next, the estimation device 10 compares each of the selected reference images with the target posting image to be processed and calculates a comparison score (S13). Note that the estimation device 10 does not execute the comparison between the non - selected reference images and the target posting image to be processed. The comparison score indicates the similarity between the two images. Here, it is assumed that the higher the similarity between the two images, the larger the comparison score calculated.

[0067] If there is a reference image whose comparison score is equal to or higher than the threshold value (Yes in S14), the estimation device 10 outputs the position information associated with that reference image as information indicating the position of the target posting image to be processed (S15).

[0068] On the other hand, when there is no reference image whose matching score is equal to or higher than the threshold value (No in S14), the estimation device 10 outputs that the position indicated by the input image to be processed is unknown (S16).

[0069] Next, another example of the processing flow of the estimation device 10 will be described using the flowchart of FIG. 6. The processing of S20 to S25 is the same as the processing of S10 to S15 described using the flowchart of FIG. 5. Hereinafter, the processing after S24 results in No will be described.

[0070] As a result of the comparison (S23) between each of the reference images selected in S22 and the input image to be processed acquired in S20, when there is no reference image whose matching score is equal to or higher than the threshold value (No in S24), the estimation device 10 compares each of the unselected reference images that were not selected in S22 with the input image to be processed (S26).

[0071] And, when there is a reference image whose matching score is equal to or higher than the threshold value (Yes in S27), the estimation device 10 outputs the position information associated with the reference image as information indicating the position of the input image to be processed (S25).

[0072] On the other hand, when there is no reference image whose matching score is equal to or higher than the threshold value (No in S27), the estimation device 10 outputs that the position indicated by the input image to be processed is unknown (S28).

[0073] Note that the unselected reference images may be grouped based on the distance between the position information associated with each reference image and the relevant area of the target poster, and the processing of S26 to S28 may be executed in order from the group with the shorter distance. And, when a reference image whose matching score is equal to or higher than the threshold value is found in a certain group, it may not be necessary to execute the processing of S26 to S28 for the remaining groups.

[0074] Also, the threshold value of S24 and the threshold value of S27 may be the same value, or may be different values. When they are different, for example, the threshold value of S24 may be larger than the threshold value of S27.

[0075] "Modification Example" The estimation device 10 may output a screen showing the related area of the target poster specified by the related area specifying unit 12. An example of such a screen is shown in FIG. 8. In the illustrated screen, the related area of the target poster is mapped on a map. Specifically, a heat map showing the related area of the target poster and a pin indicating the most relevant location are superimposed on the map. The heat map indicates the depth of relevance. The depth of relevance is specified based on the relevance score. The higher the relevance score, the deeper the relevance.

[0076] Although there are various calculation methods for the relevance score, for example, the following methods can be considered. As described above, the related area of the target poster can be specified by a plurality of methods. Therefore, the relevance score for each position is calculated according to the method by which the related area is specified. For example, the relevance score of the related area specified based on the account information of the target poster may be made higher than the relevance score of the related area specified by other methods. Also, the relevance score of the related area specified repeatedly by a plurality of methods may be made relatively high. In this case, the higher the number of repetitions, the higher the relevance score may be made.

[0077] Note that although one pin is shown in the example of FIG. 8, a plurality of pins may be displayed in association with one target poster. For example, when there are a plurality of locations where the relevance score is equal to or higher than the reference value, or when there are a plurality of locations where the difference from the highest relevance score is equal to or less than the threshold value, pins may be displayed in association with each of the plurality of locations.

[0078] By outputting such a screen, the related area of the target poster can be visually and easily provided to the user or administrator.

[0079] "Hardware Configuration" Next, an example of the hardware configuration of the estimation device 10 will be described. Each functional unit of the estimation device 10 is realized by an arbitrary combination of hardware and software centered around a CPU (Central Processing Unit), memory, program loaded into the memory, storage unit such as a hard disk storing the program (in addition to the program stored in advance at the stage of shipping the device, programs downloaded from a storage medium such as a CD (Compact Disc) or a server on the Internet can also be stored), and a network connection interface. And it is understood by those skilled in the art that there are various modifications to the realization method and device.

[0080] FIG. 7 is a block diagram illustrating the hardware configuration of the estimation device 10. As shown in FIG. 7, the estimation device 10 includes a processor 1A, a memory 2A, an input / output interface 3A, a peripheral circuit 4A, and a bus 5A. The peripheral circuit 4A includes various modules. The estimation device 10 may not have the peripheral circuit 4A. Note that the estimation device 10 may be composed of a plurality of physically and / or logically divided devices. In this case, each of the plurality of devices can have the above-described hardware configuration.

[0081] Bus 5A is a data transmission path for the processor 1A, memory 2A, peripheral circuit 4A, and input / output interface 3A to transmit and receive data with each other. The processor 1A is an arithmetic processing device such as a CPU or a GPU (Graphics Processing Unit). The memory 2A is a memory such as a RAM (Random Access Memory) or a ROM (Read Only Memory). The input / output interface 3A includes an interface for acquiring information from an input device, an external device, an external server, an external sensor, a camera, etc., and an interface for outputting information to an output device, an external device, an external server, etc. The input device is, for example, a keyboard, a mouse, a microphone, a physical button, a touch panel, etc. The output device is, for example, a display, a speaker, a printer, a mailer, etc. The processor 1A can issue commands to each module and perform operations based on their calculation results.

[0082] <Function and effect> The estimation device 10 of the present embodiment identifies the position indicated by the post image (still image and / or moving image) to be processed based on the collation result between the reference image and the post image to be processed. Then, the estimation device 10 identifies the related area of the target poster who posted the post image to be processed, and uses the reference image selected based on the related area as the target to be collated with the post image to be processed. By narrowing down the reference images to be collated in this way, the time required for processing can be shortened.

[0083] Further, the estimation device 10 identifies the related area of the target poster based on at least one of the account information of the target poster, the posting image to be processed, the past posting images of the target poster, the position information associated with the past posting images of the target poster, the posting text posted together with the posting image to be processed, the past posting texts of the target poster, the language used by the target poster, the posting timing of the target poster, the activity area of the target poster, and the related area of the related poster having a predetermined relationship with the target poster. By identifying the related area of the target poster based on such information, it is possible to accurately identify the area truly related to the target poster with high precision. As a result, the probability that the position shown in the posting image to be processed is included in the related area increases.

[0084] As described above, the embodiments of the present invention have been described with reference to the drawings, but these are examples of the present invention, and various configurations other than the above can also be adopted.

[0085] In this specification, "acquisition" means, based on user input or based on a program instruction, "the act of the own device going to retrieve data stored in another device or storage medium (active acquisition)", for example, making a request or inquiry to another device and receiving the data, accessing another device or storage medium and reading it out, etc., and, based on user input or based on a program instruction, "the act of the own device inputting data output from another device (passive acquisition)", for example, receiving data distributed (or transmitted, push-notified, etc.), and also selecting and acquiring from the received data or information, and "editing data (textualizing, rearranging data, extracting some data, changing the file format, etc.) to generate new data and acquiring the new data", including at least one of these.

[0086] Some or all of the above embodiments can also be described as follows in the appended claims, but are not limited thereto. 1. Acquisition means for acquiring a posting image to be processed posted on the Internet, Relevant area identification means for identifying the relevant area of the target poster who posted the input posted image to be processed, Selection means for selecting a reference image to be compared with the input posted image to be processed based on the identification result of the relevant area of the target poster, Position estimation means for estimating the position indicated by the input posted image to be processed based on the comparison result between the selected reference image and the input posted image to be processed, An estimation device having the above. 2. The relevant area identification means is The estimation device according to 1, which identifies the relevant area of the target poster based on the position information included in the account information of the target poster. 3. The relevant area identification means is The estimation device according to 1 or 2, which identifies the relevant area of the target poster based on at least one of the subject of the input posted image to be processed and the subject of the past posted images of the target poster. 4. The relevant area identification means is The estimation device according to any one of 1 to 3, which identifies the relevant area of the target poster based on at least one of the voice associated with the input posted image to be processed and the voice associated with the past posted images of the target poster. 5. The relevant area identification means is The estimation device according to any one of 1 to 4, which identifies the relevant area of the target poster based on the position information associated with the past posted images of the target poster. 6. The relevant area identification means is The estimation device according to any one of 1 to 5, which identifies the relevant area of the target poster based on at least one of the keyword included in the posted text posted together with the input posted image to be processed and the keyword included in the past posted texts of the target poster. 7. The relevant area identification means is The estimation device according to any one of 1 to 6, which identifies the relevant area of the target poster based on the language used by the target poster. 8. The relevant area identification means is An estimation device according to any one of 1 to 7 that specifies the relevant area of the target poster based on the posting timing of the target poster. 9. The relevant area specifying means An estimation device according to any one of 1 to 8 that specifies the relevant area of the target poster based on the activity area of the target poster. 10. The relevant area specifying means An estimation device according to any one of 1 to 9 that specifies the relevant area of the target poster based on the relevant area of a related poster having a predetermined relationship with the target poster. 11. Position information is associated with the reference image, The selection means selects the reference image associated with the position information having a predetermined relationship with the relevant area of the target poster, according to any one of 1 to 10 of the estimation device described above. 12. A computer An acquisition step of acquiring a posting image to be processed posted on the Internet, A relevant area specifying step of specifying the relevant area of the target poster who posted the posting image to be processed, A selection step of selecting a reference image to be compared with the posting image to be processed based on the specification result of the relevant area of the target poster, A position estimation step of estimating the position indicated by the posting image to be processed based on the comparison result between the selected reference image and the posting image to be processed, An estimation method for executing. 13. A computer An acquisition means for acquiring a posting image to be processed posted on the Internet, A relevant area specifying means for specifying the relevant area of the target poster who posted the posting image to be processed, A selection means for selecting a reference image to be compared with the posting image to be processed based on the specification result of the relevant area of the target poster, and A position estimation means for estimating the position indicated by the posting image to be processed based on the comparison result between the selected reference image and the posting image to be processed, A program that causes it to function as.

Description of Symbols

[0087] 10 Estimation device 11 Acquisition unit 12 Related area identification unit 13 Selection unit 14 Position estimation unit 15 Memory unit 20 Social media system molecule 1A Processor 2A Memory 3A Input / output I / F 4A Peripheral circuit 5A Bus

Claims

1. An acquisition means for acquiring a posted image to be processed posted on the Internet, A related area specifying means for specifying a related area of a target poster who posted the posted image to be processed, A selection means for selecting a reference image to be compared with the posted image to be processed based on the specification result of the related area of the target poster, A position estimation means for estimating the position indicated by the posted image to be processed based on the comparison result between the selected reference image and the posted image to be processed, An estimation device having the above.

2. The related area specifying means, The estimation device according to claim 1, wherein the related area of the target poster is specified based on position information included in the account information of the target poster.

3. The related area specifying means, The estimation device according to claim 1 or 2, wherein the related area of the target poster is specified based on at least one of a subject of the posted image to be processed and a subject of a past posted image of the target poster.

4. The related area specifying means, The estimation device according to any one of claims 1 to 3, wherein the related area of the target poster is specified based on at least one of a voice associated with the posted image to be processed and a voice associated with a past posted image of the target poster.

5. The related area specifying means, The estimation device according to any one of claims 1 to 4, wherein the related area of the target poster is specified based on position information associated with a past posted image of the target poster.

6. The related area specifying means, The estimation device according to any one of claims 1 to 5, wherein the related area of the target poster is specified based on at least one of a keyword included in a posted text posted together with the posted image to be processed and a keyword included in a past posted text of the target poster.

7. The related area specifying means, The language used by the target poster, The posting timing of the target poster, The activity area of the target poster, or The related area of a related poster having a predetermined relationship with the target poster, The estimation device according to any one of claims 1 to 6, wherein the related area of the target poster is specified based on the above.

8. Position information is associated with the reference image, The selection means selects the reference image with which the position information having a predetermined relationship with the related area of the target poster is associated. The estimation device according to any one of claims 1 to 7.

9. A computer, an acquisition step of acquiring a posted image to be processed posted on the Internet, a related area specifying step of specifying a related area of a target poster who posted the posted image to be processed, a selection step of selecting a reference image to be compared with the posted image to be processed based on the specified result of the related area of the target poster, a position estimation step of estimating the position indicated by the posted image to be processed based on the comparison result between the selected reference image and the posted image to be processed, and an estimation method for executing the above.

10. A computer, an acquisition means for acquiring a posted image to be processed posted on the Internet, a related area specifying means for specifying a related area of a target poster who posted the posted image to be processed, a selection means for selecting a reference image to be compared with the posted image to be processed based on the specified result of the related area of the target poster, and a position estimation means for estimating the position indicated by the posted image to be processed based on the comparison result between the selected reference image and the posted image to be processed, and a program for causing the computer to function as such.

Images