JP5478418B2 - Map image acquisition device, map image acquisition method, and program - Google Patents

Description

  The present invention relates to a map image acquisition device that receives one or more place name information and acquires a map image corresponding to the place name information.

  Conventionally, a map corresponding to an input keyword can be easily browsed by map distribution or the like via the Internet (see Non-Patent Document 1, for example).

“Map”, [online], July 26, 2010, [August 5, 2010 search], Wikipedia, Internet <URL: http: // ja. wikipedia. org / wiki /% E5% 9C% B0% E5% 9B% B3>

  However, there are cases where a deformed map image such as a guide map or a tourist map is more convenient than a map based on such actual measurement. On the other hand, there is a problem that it is not easy to search for a map image published on a website or the like. For example, even if you want to search for a map image of Himeji, if you perform an image search with the search keyword “Himeji”, you will find a photo image of Himeji Castle, etc. It was difficult.

  The present invention has been made to solve the above-described problems, and an object thereof is to provide a map image acquisition device or the like that can acquire a publicly available map image according to desired place name information.

To achieve the above object, a map image acquisition apparatus according to the present invention is a map that is an image including a place name information receiving unit that receives one or more place name information that is place name information, and a place name information received by the place name information receiving unit. A map image acquisition unit that acquires an image from a server that publishes an image, and an output unit that outputs a map image acquired by the map image acquisition unit.
With such a configuration, it is possible to acquire a map image including the received place name information.

  In the map image acquisition device according to the present invention, character recognition is performed by a character recognition unit that performs character recognition on a public image that is an image published on a server, and a character recognition unit included in the public image. A place name determination unit that determines whether the place name information received by the place name information reception unit is included in the character string, and the map image acquisition unit includes the place name information received by the place name information reception unit by the place name determination unit A map image that is a public image determined to be acquired may be acquired.

  With such a configuration, since character recognition is performed on a public image, even if the map image includes characters as an image, the map including the received place name information using the characters An image can be acquired.

Further, in the map image acquisition device according to the present invention, a place name that associates place name information with coordinates corresponding to the place name information as to whether or not the character string that has been character-recognized by the character recognition unit included in the public image is a place name. A threshold value determination unit for determining whether the public image includes more place names than the threshold value, and the map image acquisition unit receives the place name information by the place name determination unit. A map image that is a public image that is determined to include the place name information received by the section and that is determined to include more place names than the threshold value by the threshold determination section may be acquired.
Since an image including a place name to some extent is considered to be a map image, such a configuration can appropriately acquire a map image.

  The map image acquisition apparatus according to the present invention further includes a number specifying unit for specifying the number of one or more place name information received by the place name information receiving unit included in the map image acquired by the map image acquiring unit, and an output unit The map image may be output so that the map image having a larger number specified by the number specifying unit is ranked higher.

  Since the more place name information received is included, the map image desired by the user is considered, and thus, for example, the map image can be output in the order desired by the user. sell.

  Moreover, in the map image acquisition device according to the present invention, the place name information receiving unit receives a plurality of place name information, and includes points corresponding to the plurality of place name information attached by the place name information receiving unit. An area consistency degree calculating unit that calculates an area matching degree corresponding to the degree of overlap between the smallest area in the map and the map image acquired by the map image acquiring unit and based on the actual measurement; The output unit may output the map image so that the higher the region matching degree, the higher the ranking.

  As the overlap between the area corresponding to the received place name information and the area of the map image is larger, it is considered that the map image is desired by the user. Therefore, with such a configuration, for example, in the order desired by the user A map image can be output.

  The map image acquisition device according to the present invention further includes a spatial matching degree calculation unit that calculates a spatial matching degree indicating a degree of matching between the map image acquired by the map image acquisition unit and the map based on the actual measurement. The output unit may output the map image so that the higher the spatial matching degree, the higher the ranking.

  As the measured map and the map image are more consistent, it is considered that the map image has a higher accuracy. Therefore, with such a configuration, for example, the map images can be output in the order of higher accuracy. .

  Further, in the map image acquisition device according to the present invention, the spatial matching degree calculation unit includes a map image acquired by the map image acquisition unit from a point corresponding to one place name character string to a point corresponding to another place name character string. A spatial consistency degree indicating a correlation between a distance order and a distance order from a point corresponding to one place name character string to a point corresponding to another place name character string in a map based on actual measurement may be calculated. .

  As the actual distance is the same as the distance in the map image, the map image is considered to be a more accurate map image. With this configuration, for example, the map images are output in the order of higher accuracy. Can become possible.

  In the map image acquisition device according to the present invention, the spatial matching degree calculation unit may include other place name characters for a straight line connecting points corresponding to the first and second place name character strings in the map image acquired by the map image acquisition unit. The degree of coincidence between the position of the point corresponding to the column and the position of the point corresponding to another place name character string with respect to the straight line connecting the points corresponding to the first and second place name character strings in the map based on the actual measurement The spatial matching degree shown may be calculated.

  As the actual positional relationship and the positional relationship in the map image are the same, the map image is considered to be a highly accurate map image. With such a configuration, for example, map images can be output in order of accuracy. It can be

  Further, in the map image acquisition device according to the present invention, the peripheral text acquisition unit that acquires the peripheral text that is the text around the map image from the page that includes the map image acquired by the map image acquisition unit, and the map image acquisition unit A text matching degree calculating unit that calculates a degree of matching between the acquired map image and a place name included in the surrounding text acquired by the surrounding text acquiring unit, and the output unit includes the text matching degree. The map image may be output so that the higher the ranking, the higher the ranking.

  Because the map image and the surrounding text of the map image are more consistent, it is considered that the map image is carefully created according to the surrounding text, and is considered to be a high-quality map image With such a configuration, for example, map images can be output in descending order of quality.

  Moreover, in the map image acquisition device according to the present invention, the text matching degree calculation unit generates a place name database that associates place name information and coordinates corresponding to the place name information among character strings included in the map image acquired by the map image acquisition unit. The text matching degree may be calculated according to the degree of matching between the place name character string, which is a character string determined to be a place name, and the place name included in the surrounding text.

  As the place name character string included in the map image and the place name included in the surrounding text match, it is considered that the map image is carefully created according to the surrounding text, and the map image is of high quality. Since it is considered that there is, it is possible to output map images in descending order of quality, for example, with such a configuration.

  In the map image acquisition device according to the present invention, the text matching degree calculation unit includes the minimum area in the map based on the actual measurement including the point corresponding to the place name included in the surrounding text, and the map acquired by the map image acquisition unit. The text matching degree corresponding to the degree of overlap with the map area based on the actual measurement corresponding to the image may be calculated.

  The larger the overlap between the map image area and the area corresponding to the place name included in the surrounding text, the more carefully the map image is considered to be created according to the surrounding text, and the map image has a higher quality. Therefore, with such a configuration, for example, map images can be output in descending order of quality.

  According to the map image acquisition device and the like according to the present invention, a map image including the received place name information can be acquired.

The block diagram which shows the structure of the map image acquisition apparatus by Embodiment 1 of this invention. The flowchart which shows operation | movement of the map image acquisition apparatus by the embodiment The flowchart which shows operation | movement of the map image acquisition apparatus by the embodiment The flowchart which shows operation | movement of the map image acquisition apparatus by the embodiment The figure which shows an example of the system containing the map image acquisition apparatus by the embodiment The figure which shows an example of the place name database in the embodiment The figure which shows an example of the acquired image, URL, etc. in the embodiment The figure which shows an example of the map image in the embodiment The figure which shows an example of the map image in the embodiment The figure which shows an example of the acquired image, URL, etc. in the embodiment The figure which shows an example of the acquired image, URL, etc. in the embodiment The figure which shows an example of the acquired image, URL, etc. in the embodiment The figure which shows an example of the score information in the embodiment The figure for demonstrating calculation of the area | region consistency degree in the embodiment The figure for demonstrating calculation of the area | region consistency degree in the embodiment The figure which shows an example of the displayed map image in the embodiment Schematic diagram showing an example of the appearance of the computer system in the embodiment The figure which shows an example of a structure of the computer system in the embodiment

  Hereinafter, a map image acquisition device according to the present invention will be described using embodiments. In the following embodiments, components and steps denoted by the same reference numerals are the same or equivalent, and repetitive description may be omitted.

(Embodiment 1)
A map image acquisition apparatus according to Embodiment 1 of the present invention will be described with reference to the drawings. The map image acquisition apparatus according to the present embodiment acquires a map image including the place name of the received place name information.

  FIG. 1 is a block diagram showing a configuration of a map image acquisition apparatus 1 according to this embodiment. The map image acquisition device 1 according to the present embodiment includes a place name information receiving unit 11, a character recognition unit 12, a place name determining unit 13, a threshold value determining unit 14, a map image acquiring unit 15, and a number specifying unit 16. A region matching degree calculation unit 17, an actual measurement map information acquisition unit 18, a spatial matching degree calculation unit 19, a surrounding text acquisition unit 20, a text matching degree calculation unit 21, and an output unit 22.

  The place name information receiving unit 11 receives one or more place name information. The place name information is place name information, and any name can be used as long as the name corresponds to any position on the map. The place name information may be, for example, the name of the land (eg, “Himeji”, “Kobe”, etc.), the name of a famous place (eg, “Sekigahara Battlefield”, etc.), or the name of the building ( For example, it may be “Himeji Castle”, “Tsutenkaku”, etc.), or may be a name related to transportation (eg, “Himeji Station”, “Kobe Station”, etc.) Or “Himeji City Museum of Art”). The place name information receiving unit 11 may receive one piece of place name information or a plurality of place name information.

  The place name information receiving unit 11 may receive, for example, one or more place name information input from an input device (for example, a keyboard, a mouse, a touch panel, etc.), and one or more sent via a wired or wireless communication line. The place name information may be received, or one or more place name information read from a predetermined recording medium (for example, an optical disk, a magnetic disk, or a semiconductor memory) may be received. Note that the place name information receiving unit 11 may or may not include a device (for example, a modem or a network card) for receiving. The place name information receiving unit 11 may be realized by hardware, or may be realized by software such as a driver that drives a predetermined device.

  The character recognition unit 12 performs character recognition on a public image that is an image publicized on the server. The public image may be acquired by the map image acquisition unit 15 or may be acquired by another component. In the present embodiment, it is assumed that the map image acquisition unit 15 acquires a public image. Character recognition is the conversion of a character image into a character code. Character recognition is already known and will not be described in detail. A character string that is a result of character recognition performed by the character recognition unit 12 (this character string is indicated by a character code) may be stored in a recording medium (not shown). The position of the character image corresponding to the character string may also be stored on a recording medium (not shown).

  The place name determination unit 13 determines whether or not the place name information received by the place name information reception unit 11 is included in one or more character strings that have been subjected to character recognition by the character recognition unit 12 included in the public image. For example, when the place name information receiving unit 11 receives the place name information “Himeji” and the character recognition unit 12 recognizes the character strings “Himeji”, “Akashi”, and “Kakogawa”, the place name determining unit 13 sets the place name information “ It is determined that “Himeji” is included in the character string. When the place name information receiving unit 11 receives a plurality of pieces of place name information, the place name determining unit 13 includes at least one place name information among the plurality of place name information in one or more character strings in which character recognition has been performed. The place name information may be determined to be included, or the place name information is determined to be included when all of the plurality of place name information are included in one or more character-recognized character strings. May be. In the present embodiment, the former case will be described.

  The threshold value determination unit 14 determines whether or not the character string that has been subjected to character recognition by the character recognition unit 12 included in the public image is a place name. The threshold determination unit 14 may make the determination by accessing the place name database. The place name database is a database that associates place name information with coordinates corresponding to the place name information. Therefore, the threshold value determination unit 14 can determine that the character string is a place name when the character string subjected to character recognition is place name information associated with the coordinates in the place name database. . The coordinates may be, for example, latitude / longitude, or other coordinates based on a certain reference point. Further, the threshold determination unit 14 determines whether or not the public image includes more place names than the threshold. The “place names greater than the threshold value” may be a place name having a larger number than the threshold value, or may be a place name having a ratio higher than the threshold value. In the former case, for example, if the threshold value is “3” and it is determined that four character strings are place names among character strings recognized as characters, the public image has more than the threshold value. The place name will be included. In the latter case, for example, it is determined that the threshold value is “0.5”, the number of character strings that have been recognized is seven, and four of the character strings are place names. , More place names than the threshold value are included in the public image. Note that “more than a threshold value” may include a threshold value or may not include a threshold value.

  The map image acquisition unit 15 acquires a map image that is an image including the place name information received by the place name information receiving unit 11 from a server that publishes the image. In the present embodiment, the map image acquisition unit 15 first acquires a public image. The public image may be stored in a recording medium (not shown). In order to facilitate the acquisition of the surrounding text by the surrounding text acquisition unit 20, the map image acquisition unit 15 includes information for accessing a page including the public image (for example, the page of the page). URL or the like) may be stored in association with the public image. Then, the map image acquiring unit 15 determines that the place name information received by the place name information receiving unit 11 is included in the public image by the place name determining unit 13 and the threshold value determining unit 14 exceeds the threshold value. When it is determined that the place name is included, a map image that is the public image is acquired. By acquiring a public image determined to include place name information as a map image, the map image acquisition unit 15 acquires a map image including place name information. Further, since an image including many place names is considered to be a map image, the map image acquiring unit 15 obtains a map image that is a public image including more place names than a threshold value. The possibility of acquiring a public image increases.

  In addition, when acquiring a public image, the map image acquisition part 15 may acquire all the public images, or the image search which used one or more place name information which the place name information reception part 11 received as a search key You may acquire the public image which hit by. The image search is already known and will not be described in detail. The public image may be an image published on a website on the Internet, or may be an image published on an intranet or a local database. Further, the process of acquiring the map image may be, for example, accumulating the map image in a predetermined recording medium (not shown), or among the public images stored in the recording medium (not shown) in advance. A flag indicating that the map image is a map image may be set.

  The number specifying unit 16 specifies the number of one or more place name information received by the place name information receiving unit 11 included in the map image acquired by the map image acquiring unit 15. The number specified by the number specifying unit 16 may be stored in a recording medium (not shown). When the place name determination unit 13 determines that place name information is included when all the received place name information is included in one or more character strings that have been character-recognized, the number specifying unit 16 specifies the place name information. The number of place name information to be made is equal to the number of place name information received. Therefore, when the number of place name information is specified by the number specifying unit 16, the place name determining unit 13 is one in which at least one place name information among the received one or more place name information is character-recognized. When included in the above character string, it is preferable to determine that the place name information is included.

The region matching degree calculation unit 17 includes a minimum region (hereinafter referred to as “region A”) in a map based on actual measurements including points corresponding to a plurality of place name information attached by the place name information receiving unit 11, The degree of area matching corresponding to the degree of overlap with the map area (hereinafter referred to as “area B”) based on the actual measurement corresponding to the map image acquired by the image acquisition unit 15 is calculated. In this case, it is assumed that the place name information receiving unit 11 receives a plurality of place name information. If a set of a plurality of place name information received by the place name information receiving unit 11 is Q, the area A = MBR (Q). The MBR is a function that returns a minimum area in a map based on actual measurement including points corresponding to a plurality of place name information. Each place name information included in Q can be converted into coordinates by using the place name database described above. And the smallest area | region becomes area | region A = MBR (Q) among the rectangular area | regions where all the coordinates are included and an east-west direction and a north-south direction are the directions of each side. This area A is an area on the coordinate plane corresponding to the coordinates of the place name database. In the case where the coordinates are latitude / longitude, for example, the region A has the smallest latitude, the largest latitude, the smallest longitude, and the largest among all the coordinates corresponding to the local name information included in Q. A rectangle surrounded by four straight lines corresponding to a large longitude. Further, if the map image corresponding to the region B is m and the set of place names included in the map image m is M, the region B = EBR (M, m). Note that EBR (M, m) is a function that returns a map area corresponding to the map image m based on actual measurement. Usually, since a place name is included in the map image m, in the map image m, the position of all the place names is included, and the smallest area among rectangular areas whose vertical and horizontal directions of the map image m are directions of the respective sides. Is region C. This area C is an area on the map image m. The position of the place name may be the center position of the place name character string, the position of the predetermined outer edge of the place name character string (for example, upper left), or all the place name character strings. May be the smallest rectangular area including Further, the area MBR (M) corresponding to M can be obtained using the function MBR. And the area | region which expanded the area | region MBR (M) in the same ratio as extending the area | region C to the whole map image m becomes area | region B = EBR (M, m). For example, when the area C is multiplied by α upward, β times downward, γ times right, and δ times left, and if it becomes the same size as the entire map image m, MBR (M) is set upward. Region B is obtained by multiplying by α, by β by downward, by γ by right, and by δ by left. Here, it is assumed that the direction is the same in the map image m and the coordinate plane. Α, β, γ, and δ are each a real number of 1 or more. If the function that returns the area of a certain region X is AS (X), the region matching degree R is expressed by the following equation.
Region consistency R
= AS (MBR (Q) ∩EBR (M, m)) / AS (MBR (Q) ∪EBR (M, m))

  The actual measurement map information acquisition unit 18 acquires actual map information that is map information based on the actual measurement. A map based on actual measurement is a map showing the same positional relationship as the real world positional relationship or the same distance relationship as the real world distance relationship. The map based on the actual measurement may be a survey map which is a map actually created by surveying, or may be a map substantially equivalent to the map, for example, a map created from aerial photographs. Since the actual map information acquisition unit 18 can acquire the actual map information in the same manner as the normal online map acquisition, the detailed description thereof is omitted. The actual map information acquisition unit 18 may acquire the actual map information via a network such as the Internet, or may acquire the actual map information by reading the actual map information stored locally. Also good. The measured map information acquisition unit 18 includes a place name character string that is a character string determined to be a place name using the place name database described above, among the character strings included in the map image acquired by the map image acquisition unit 15. Acquire measured map information. The actual map information acquisition unit 18 may acquire actual map information including one or more place name character strings by performing a search using one or more place name character strings as a search key, for example. The acquired actual map information may be stored in a recording medium (not shown).

  The spatial matching degree calculation unit 19 calculates a spatial matching degree indicating the degree of matching between the map image acquired by the map image acquisition unit 15 and the actual map information acquired by the actual map information acquisition unit 18. In the present embodiment, the spatial matching degree calculation unit 19 calculates the spatial matching degree S1 according to the correlation of the order of distances between the two and the spatial matching degree S2 according to the degree of coincidence of the positional relationship between the points in both. The case where it does is demonstrated.

First, the spatial matching degree S1 corresponding to the correlation of the order of distances between the map image and the measured map information will be described. The spatial matching degree calculation unit 19 obtains the order of distances from a point corresponding to one place name character string to a point corresponding to another place name character string in the map image acquired by the map image acquisition unit 15. The method of selecting the one place name character string does not matter. For example, a place name character string near the center of the map image may be selected, or a place name character string that is a starting place or a destination place in the map image may be selected. For the method of specifying the place name character string of the departure place and the place name character string of the destination in the map image, refer to the following document 1.
Reference 1: Daisuke Kitayama, Lilong, Kazutoshi Kakutani, “Authenticity analysis based on intention extraction for deformed maps”, DEIM 2010, February 2010

  When the one place name character string is selected, the spatial matching degree calculation unit 19 calculates the distance between the one place name character string and the other place name character strings, and sorts them according to the distance. As a result, it is possible to know the order of the distance between the reference place name character string and other place name character strings. The sort may be in descending order or ascending order.

  In addition, the space matching degree calculation unit 19 calculates the distance between the point corresponding to the one place name character string and the point corresponding to another place name character string in the actually measured map information acquired by the actually measured map information acquisition unit 18. Calculate and sort according to the distance. As a result, it is possible to know the order of the distance between the reference place name character string and other place name character strings. The sort is assumed to be the same sort as that for the map image.

Thereafter, the spatial consistency calculation unit 19 calculates a spatial consistency S1 indicating a correlation between the order of distances in the map information and the order of distances in the actually measured map information. The spatial matching degree may be, for example, Spearman's rank correlation coefficient expressed by the following equation. In general, Spearman's rank correlation coefficient takes a value from −1 to 1, but takes a value between 0 and 1 in the following equation. In the following equation, D is the difference between the order in the map information regarding a certain place name character string (which is not a reference place name character string) and the order in the measured map information. N is the number of distances obtained. It goes without saying that the spatial matching degree S1 may be calculated using other than Spearman's rank correlation coefficient.

Next, the spatial matching degree S2 corresponding to the degree of coincidence of the positional relationship between points in the map image and the measured map information will be described. The spatial matching degree calculation unit 19 includes a position of a point corresponding to another place name character string relative to a straight line connecting points corresponding to the first and second place name character strings in the map image acquired by the map image acquisition unit 15; In the actual map information acquired by the actual map information acquisition unit 18, the degree of coincidence with the position of a point corresponding to another place name character string with respect to a straight line connecting the points corresponding to the first and second place name character strings described above. The spatial matching degree S2 shown is calculated. In particular,
_{S2 = (Σ i c (o} 1, o 2, o i)) / k
It becomes. k is the number of place name character strings other than the first and second place name character strings. O ₁ is a first place name character string, o ₂ is a second place name character string, and o _i is an i th place name character string other than the first and second place name character strings. is there. Further, c (o ₁ , o ₂ , o _i ) has a position corresponding to o _i with respect to a straight line connecting the positions of o ₁ and o _{2 in} both the map image and the measured map information. The function is 1 when it exists on the same side and 0 when it exists on a different side. Whether they are on the same side may be determined as follows, for example. vector V1 to position end point corresponding to the position corresponding to the o ₁ to o ₂ as a starting point, the vector V2 Metropolitan to ending the position corresponding to the o _i a position corresponding to the o ₁ as the starting point, from the vector V1 Calculate the clockwise angle to vector V2. If the angle is in the range of 0 degrees or more and less than 180 degrees in both the map image and the measured map information, c (o ₁ , o ₂ , o _i ) = 1. If the angle is in the range of 180 degrees or more and less than 360 degrees in both the map image and the measured map information, c (o ₁ , o ₂ , o _i ) = 1. Also, if the angle is in the range of 0 degree or more and less than 180 degree on one of the map image and the measured map information and on the other hand, if the angle is in the range of 180 degree or more and less than 360 degree, c (o ₁ , o ₂ , o _i ) = 0.

In addition, the method by which the space matching degree calculation unit 19 selects the first and second place name character strings does not matter. For example, two place name character strings that are in a positional relationship facing the center of the map image may be selected. Further, the space matching degree calculation unit 19 may calculate an average of the above S2 for all combinations related to o ₁ and o ₂ as the space matching degree S2. In this case, all pairs of place name character strings become the first and second place name character strings.

  The peripheral text acquisition unit 20 acquires peripheral text, which is text around the map image, from a page including the map image acquired by the map image acquisition unit 15. The surrounding text may be all the text included in the page including the map image, or may be the text included in the range from the map image to the nearest heading. When the page is described in HTML, the surrounding text acquisition unit 20, for example, from the map image to the text in the range from the map image to the most recent heading tag before the map image and the map image, The text in the range up to the most recent heading tag may be acquired as the surrounding text. The surrounding text acquisition unit 20 may acquire only the text up to the heading on the near side of the map image as the surrounding text, or may acquire only the text up to the heading on the back side of the map image as the surrounding text. . The peripheral text acquired by the peripheral text acquisition unit 20 may be stored in a recording medium (not shown).

  The text matching degree calculating unit 21 calculates a text matching degree indicating the degree of matching between the map image acquired by the map image acquiring unit 15 and the place name included in the surrounding text acquired by the surrounding text acquiring unit 20. In the present embodiment, the text matching degree calculation unit 21 uses the text matching degree T1 corresponding to the degree of matching between the place name character string included in the map image and the place name included in the surrounding text, and the area corresponding to the map image. A description will be given of a case where the text matching degree T2 is calculated according to the degree of overlap with the minimum area including the point corresponding to the place name included in the surrounding text.

First, the text matching level T1 corresponding to the degree of matching between the place name character string included in the map image and the place name included in the surrounding text will be described. The text matching degree calculation unit 21 acquires a place name character string that is a character string determined to be a place name using the above-described place name database among the character strings included in the map image acquired by the map image acquisition unit 15. . A set of the place name character strings is SP1. Further, the text matching degree calculation unit 21 acquires a place name included in the surrounding text. For example, the text matching degree calculation unit 21 may perform morphological analysis on the surrounding text and acquire a proper noun that is a place name as the place name. For example, when morphological analysis is performed using MeCab, morphemes classified as “nouns, proper nouns, and regions” may be acquired as place names. Alternatively, the text matching degree calculation unit 21 may acquire a place name that matches the place name included in the place name database described above as a place name. A set of place names acquired from the surrounding text is SP2. Then, the text matching degree calculation unit 21 may calculate the text matching degree T1 by the following equation. Note that CU (X) is a function that returns the number of elements in the set X.
T1 = CU (SP1∩SP2) / CU (SP2)

Note that the text matching degree T1 may be calculated by the following equation, or by another equation that indicates a value corresponding to the degree of matching between the place name character string included in the map image and the place name included in the surrounding text. May be.
T1 = CU (SP1∩SP2) / CU (SP1∪SP2)

Next, the text matching degree T2 according to the degree of overlap between the area corresponding to the map image and the minimum area including the point corresponding to the place name included in the surrounding text will be described. The text matching degree calculation unit 21 corresponds to the minimum region MBR (SP2) in the map based on the actual measurement including the point corresponding to the place name included in the surrounding text, and the map image m acquired by the map image acquisition unit 15. Then, the text matching degree T2 corresponding to the degree of overlap with the map area EBR (SP1, m) based on the actual measurement is calculated. MBR and EBR are as described above. The text matching degree T2 may be expressed by the following equation, for example.
T2 = min (1, AS (MBR (SP2)) / AS (EBR (SP1, m)))

Note that the text matching degree T2 may be calculated by the following equation, and depends on the degree of overlap between the area corresponding to the map image and the minimum area including the point corresponding to the place name included in the surrounding text. It may be calculated by an expression indicating the value.
T2 = AS (MBR (SP2) ∩EBR (SP1, m)) / AS (MBR (SP2) ∪EBR (SP1, m))

  The output unit 22 outputs the map image acquired by the map image acquisition unit 15. Note that the output unit 22 outputs the map image so that the map image with the larger number N specified by the number specifying unit 16 is ranked higher. The output unit 22 outputs a map image so that the higher the region matching degree R, the higher the ranking. Further, the output unit 22 outputs a map image so that it ranks higher as the spatial matching degree S1 is higher, and outputs a map image so that it ranks higher as the spatial matching degree S2 is higher. The output unit 22 outputs a map image so that the higher the text matching level T1 is, the higher the ranking is, and the higher the text matching level T2, the higher the ranking is. Note that the output of the map image so that the output unit 22 ranks higher as X increases means that the ranking is performed in descending order of X when conditions other than X are the same. X is N, R, S1, S2, T1, T2, or the like. Therefore, the output unit 22 uses N, R, S1, S2, T1, and T2 as arguments, calculates a score that is an increasing function for each argument, and the map image ranks higher as the score increases. May be output. For example, the output unit 22 outputs a map image so that the higher the X, the higher the ranking is. For example, the higher the X is, the higher the output (the front) is to display the map image. Alternatively, it may be output in the order of higher X, or higher X may be output in association with higher ranking values. That is, the output of the map image may be, for example, an output in ranking order, or may be an output in which the map image is associated with the ranking value of the map image.

  Here, the output may be, for example, display on a display device (for example, a CRT or a liquid crystal display), transmission via a communication line to a predetermined device, printing by a printer, or output to a recording medium. It may be accumulated or delivered to another component. In the present embodiment, a case where the output unit 22 displays map images in order of ranking will be described. The output unit 22 may or may not include an output device (for example, a display device or a printer). The output unit 22 may be realized by hardware, or may be realized by software such as a driver that drives these devices.

Next, operation | movement of the map image acquisition apparatus 1 by this Embodiment is demonstrated using the flowchart of FIG.
(Step S101) The place name information accepting unit 11 determines whether one or more place name information has been accepted. If one or more place name information is added, the process proceeds to step S102, and if not, the process of step S101 is repeated until it is accepted. The received one or more place name information may be stored in a recording medium (not shown).

  (Step S102) The map image acquisition unit 15 acquires a public image. As described above, the map image acquisition unit 15 may acquire all public images, or perform an image search using one or more received place name information as a search key to acquire a hit public image. May be. Note that a predetermined number (for example, 100 or 1000) of public images may be acquired in descending order of the degree of coincidence with the search key that is one or more place name information among the public images that have been hit. . The acquired public image may be stored on a recording medium (not shown).

  (Step S103) The map image acquisition unit 15 or the like performs processing for acquiring a map image. Details of this processing will be described later with reference to the flowchart of FIG.

  (Step S104) The number specifying unit 16 and the like specify the number of place name information included in the map image, calculate the degree of region matching, and the like. Details of this processing will be described later with reference to the flowchart of FIG.

(Step S105) The output unit 22 ranks and outputs the map images acquired by the map image acquisition unit 15 according to the degree of region matching or the like. Then, the process returns to step S101.
In the flowchart of FIG. 2, the process is terminated by powering off or a process termination interrupt.

FIG. 3 is a flowchart showing details of the map image acquisition process (step S103) in the flowchart of FIG.
(Step S201) The map image acquisition unit 15 sets the counter i to 1.

  (Step S202) The character recognizing unit 12 performs character recognition on the i-th public image and accumulates the character string of the recognition result in a recording medium (not shown). In addition, it is preferable to accumulate so that it can be understood that the character string corresponds to the i-th public image.

  (Step S203) The place name determination unit 13 determines whether the place name information received by the place name information reception unit 11 is included in the character string recognized by the character recognition unit 12 for the i-th public image. If the place name information is included, the process proceeds to step S204. If not, the process proceeds to step S207. In this determination, for example, when one character string matches one place name information, it may be determined that the recognized character string includes the place name information, or one character string includes one place name information. If the place name information is included, it may be determined that the place name information is included in the recognized character string.

  (Step S204) The threshold value determination unit 14 specifies a place name character string included in the character string recognized by the character recognition unit 12 for the i-th public image. As described above, this specification may be performed by a method of searching a place name database using a character string recognized as a search key and determining that the character string is a place name when a hit is found. The hit may be, for example, a perfect match hit or a partial match hit. Then, the character string determined to be the place name character string may be accumulated in a recording medium (not shown) in which the place name character string is stored, or a flag indicating the place name character string is set, for example. May be. The place name character string specified in this way may be used in later processing (for example, processing for calculating the degree of region matching).

  (Step S205) The threshold value determination unit 14 reads a threshold value stored in advance on a recording medium (not shown), and determines whether or not the i-th public image includes more place name character strings than the threshold value. If more place name character strings than the threshold are included, the process proceeds to step S206. Otherwise, the process proceeds to step S207. Note that “more than the threshold value” may mean that the number is large or the ratio is large as described above.

  (Step S206) The map image acquisition unit 15 acquires the i-th public image as a map image. This acquisition may be, for example, accumulating the i-th public image in a recording medium (not shown) that stores the map image, or the flag indicating that the map image is the i-th public image. It may be set to.

  (Step S207) The map image acquisition unit 15 increments the counter i by one.

  (Step S208) The map image acquisition unit 15 determines whether or not the i-th public image exists in the public image acquired in step S102. And when it exists, it returns to step S202, and when that is not right, it returns to the flowchart of FIG.

FIG. 4 is a flowchart showing details of the processing for calculating the degree of consistency and the like (step S104) in the flowchart of FIG.
(Step S301) The output unit 22 sets the counter i to 1.

  (Step S302) The number specifying unit 16 specifies the number N of one or more accepted place name information included in the i-th map image. For example, the number identification unit 16 searches the character string recognized by the character recognition unit 12 for the i-th map image using the received local name information as a search key. The number N may be specified by repeatedly executing the process of determining that the place name information is included in the i-th map image for all the place name information. In that case, the number of hits is N. The specified number N may be stored in a recording medium (not shown).

  (Step S303) The region matching degree calculation unit 17 determines whether the place name information receiving unit 11 has received two or more place name information. If two or more place name information is received, the process proceeds to step S304, and if not, the process proceeds to step S305. This is because the region matching degree R can be calculated only when two or more pieces of place name information are received.

  (Step S304) The region matching degree calculation unit 17 obtains a region MBR (Q) corresponding to the accepted set Q of two or more place name information. Further, the region matching degree calculation unit 17 obtains a map region EBR (M, m) based on the actual measurement corresponding to the i-th map image m. M is a set of place name character strings included in the i-th map image m. Then, the region matching degree calculation unit 17 calculates a region matching degree R according to the degree of overlap between the region MBR (Q) and the region EBR (M, m). As the calculation formula of the region matching degree R, the above-described formula may be used. The calculated region matching degree R may be stored in a recording medium (not shown).

  (Step S305) The actual map information acquisition unit 18 acquires actual map information including all the place name character strings included in the i-th map image. The acquired actual map information may be stored in a recording medium (not shown).

  (Step S306) The spatial matching level calculation unit 19 obtains the order of distances from a point corresponding to one place name character string to a point corresponding to another place name character string in the i-th map image. In addition, the spatial matching degree calculation unit 19 determines the order of the distance from the point corresponding to the one place name character string to the point corresponding to the other place name character string in the measured map information acquired by the measured map information acquisition unit 18. Ask. Then, the spatial matching degree calculation unit 19 calculates a spatial matching degree S1 indicating the degree of matching between the two orders. The Spearman rank correlation coefficient described above may be used as a formula for calculating the spatial matching degree S1. The calculated spatial matching degree S1 may be stored in a recording medium (not shown).

  (Step S307) The spatial matching degree calculation unit 19 determines the position of the point corresponding to another place name character string (the straight line corresponding to the line connecting the points corresponding to the first and second place name character strings in the i-th map image. Which side is relative to). In addition, the spatial matching degree calculation unit 19 is a point corresponding to another place name character string with respect to a straight line connecting points corresponding to the first and second place name character strings in the actually measured map information acquired by the actually measured map information acquisition unit 18. Is specified (which side is the straight line). Then, the space matching degree calculation unit 19 has the same place name character string in the i-th map image and the measured map information with respect to the straight line connecting the points corresponding to the first and second place name character strings. The spatial matching degree S2 indicating the degree of coincidence of whether it is on the side or on the opposite side is calculated. The above formula may be used as a formula for calculating the degree of spatial matching S2. The calculated spatial matching degree S2 may be stored in a recording medium (not shown).

  (Step S308) The surrounding text acquisition unit 20 acquires surrounding text that is text around the i-th map image from a page including the i-th map image. For example, when the page is a web page and associated with the i-th map image and the URL of the web page including the i-th map image is stored, the surrounding text acquisition unit 20 By using the URL corresponding to the i-th map image, it is possible to access a page including the i-th map image, and to obtain surrounding text from the page. Note that the acquired peripheral text may be stored in a recording medium (not shown).

  (Step S309) The text matching degree calculation unit 21 acquires a set SP2 of place names from the surrounding text corresponding to the i-th map image. Then, the text matching degree calculation unit 21 uses the place name character string set SP1 included in the i-th map image and the acquired place name set SP2, and the place name character string included in the i-th map image, A text matching level T1 corresponding to the degree of matching with the place name included in the surrounding text is calculated. The above formula may be used as a calculation formula for the text matching degree T1. The calculated text matching degree T1 may be stored in a recording medium (not shown).

  (Step S310) The text matching degree calculation unit 21 obtains a map area EBR (SP1, m) corresponding to the i-th map image m based on actual measurement. In addition, the text matching degree calculation unit 21 includes a minimum region MBR (in a map based on actual measurement including a point corresponding to each place name included in the set of place names SP2 included in the surrounding text corresponding to the i-th map image ( SP2) is obtained. Then, the text matching degree calculation unit 21 uses the area EBR (SP1, m) and the area MBR (SP2) to calculate the text matching degree T2 according to the degree of overlap between the two. The above formula may be used as a calculation formula for the text matching degree T2. The calculated text matching degree T2 may be stored in a recording medium (not shown).

  (Step S311) The output unit 22 calculates a score using an increasing function of each argument with the number N, the region consistency R, the space consistency S1, S2, and the text consistency T1, T2 as arguments. The calculated score may be stored in a recording medium (not shown). If the region matching degree R is not calculated, the score may be calculated with R = 0.

  (Step S312) The output unit 22 increments the counter i by 1.

  (Step S313) The output unit 22 determines whether or not the i-th map image exists in the map image acquired by the map image acquisition unit 15. If the i-th map image exists, the process returns to step S302; otherwise, the process returns to the flowchart of FIG.

  Although the case where the output unit 22 manages the counter i has been described in the flowchart of FIG. 4, it goes without saying that another component may manage the counter i. Further, if the map image does not include three or more place name character strings, the significant spatial matching degree S1 cannot be calculated. Further, if the map image does not include four or more place name character strings, the significant spatial matching degree S2 cannot be calculated. Further, if the map image includes two or more place name character strings and the surrounding text does not contain two or more place names, the text matching degree T2 cannot be calculated. Therefore, when the significant spatial matching degrees S1 and S2 cannot be calculated, or when the text matching degree T2 cannot be calculated, it is not necessary to calculate them.

  Next, operation | movement of the map image acquisition apparatus 1 by this Embodiment is demonstrated using a specific example. In this specific example, as shown in FIG. 5, a case where the map image acquisition apparatus 1 is connected to the Internet 500 and acquires a map image from the web server 2 via the Internet 500 will be described. It is assumed that the map image acquisition device 1 has a place name database inside the device. The place name database is shown in FIG. 6, and place names and coordinates are associated with each other. The coordinates are assumed to be latitude / longitude. For example, since the coordinates (X101, Y101) correspond to the place name “Kobe”, it can be seen that the latitude / longitude of Kobe is (X101, Y101). In addition, by using this place name database, it can be seen that, for example, Kobe and Akashi are place name character strings.

  First, it is assumed that the user inputs place name information “Himeji” and “Akashi” by operating the keyboard and mouse of the map image acquisition device 1. Then, the place name information is received by the place name information receiving unit 11 and stored in a recording medium (not shown) (step S101). In addition, the place name information reception unit 11 passes to the map image acquisition unit 15 that the place name information has been received. When the map image acquisition unit 15 receives the fact that the place name information has been received, the map image acquisition unit 15 reads the place name information “Himeji” and “Akashi” from a recording medium (not shown), and obtains a public image by performing an image search using these as search keys. (Step S102). As shown in FIG. 7, these acquired images are stored in a recording medium (not shown) in association with the URL of the web page in which the image is included. In FIG. 7, an ID for identifying a record, a public image, a URL of a web page including the public image, character string information, and a map image flag are associated with each other. For example, it is understood that the image “ABC-MAP.gif” is included in a web page whose URL is “http: //...abc-u.ac.jp”. The image “ABC-MAP.gif” is assumed to be as shown in FIG. In addition, it is assumed that the image “BRIDGE.jpg” is a picture of the Akashi Kaikyo Bridge and not a map. The image “XYZ-MAP.jpg” is assumed to be the one shown in FIG. The character string information and the map image flag will be described later.

  Thereafter, when the map image acquisition unit 15 passes the fact that the public image has been acquired to the character recognition unit 12 or the like, the process of acquiring the map image is started (step S103). Specifically, the character recognizing unit 12 performs character recognition on the image “ABC-MAP.gif” shown in FIG. 8, and character strings “Kobe”, “Akashi”, “Himeji”, “ABC University”, “Walk 15”. "Minute" is recognized (steps S201 and S202). Each character string and the position of the character string in the map image are associated with the image “ABC-MAP.gif” as indicated by the character string and position included in the character string information of FIG. accumulate. The position of the character string is the position of a specific part of the character string. For example, the position may be the center position of the character string or the upper left position of the character string. The position of the character string is a position in the map image, and may be, for example, the number of pixels in the right direction and the downward direction with the upper left corner of the map image as a base point.

  Next, the place name determining unit 13 determines whether the place name information received by the place name information receiving unit 11 is included in the recognized character string (step S203). Specifically, the place name determination unit 13 first searches for a character string corresponding to the image “ABC-MAP.gif” using the received place name information “Himeji” as a search key. In this case, since it is a hit, the place name determination unit 13 determines that the place name information is included in the public image of ID1. Thereafter, the threshold judgment unit 14 searches the place name database using each recognized character string as a search key, sets the place name flag “1” for the hit character string, and sets the character string that has not been hit. Is set with the place name flag “0” (step S204). The place name flag “1” is a flag indicating that the character string corresponding to the place name flag is a place name character string, and the place name flag “0” indicates that the character string corresponding to the place name flag is not a place name character string. Therefore, setting the place name flag is processing for specifying the place name character string. In the character string information shown in FIG. 10, a character string that has been character-recognized, a position in the image of the character string, and a place name flag indicating whether the character string is a place name are associated with each other. Further, the threshold determination unit 14 determines whether or not the place name character string is equal to or greater than the threshold (step S205). Here, in this specific example, the threshold value is set to “3”, and the threshold value determination unit 14 determines that the threshold value is greater than the threshold value when three or more place name character strings are included in the image. Assume that many place name character strings are included. Therefore, in this case, the threshold value determination unit 14 counts the number “3” of character strings whose place name flag corresponding to the image “ABC-MAP.gif” is “1”, which is the threshold value. Therefore, it is determined that the image “ABC-MAP.gif” includes a place name character string that is equal to or greater than the threshold value, and the fact is transferred to the map image acquisition unit 15. Then, the map image acquisition unit 15 sets the map image flag corresponding to the image “ABC-MAP.gif” to “1” as shown in FIG. 11 (step S206). Here, the map image flag “1” is a flag indicating that the image is a map image, and the map image flag “0” is a flag indicating that the image is not a map image. Therefore, the process of setting the map image flag “1” is a process of acquiring a map image. The map image acquisition unit 15 performs map image flag “0” for an image determined not to include the place name information or an image determined not to include more place name character strings than the threshold value. Shall be set.

  The character recognizing unit 12 and the like repeatedly execute the same processing for other acquired images (steps S202 to S208). Since the image “BRIDGE.jpg” is a picture of the Akashi Kaikyo Bridge, no character string is included, and the map image flag “0” is set. Further, since the image “XYZ-MAP.jpg” shown in FIG. 9 includes the place name character strings “Himeji”, “Kakogawa”, etc., the map image flag “1” is set. As a result, the character string information and the map image flag are as shown in FIG.

  Next, the number specifying unit 16 and the like perform processing such as specifying the number of place name information and calculating the degree of region matching, etc., for the map image, that is, the image for which the map image flag “1” is set (step S104). ). Specifically, the number specifying unit 16 searches for a character string corresponding to the first map image using each of the received place name information “Himeji” and “Akashi” as search keys. Then, the hit number “2” is accumulated as the number N of place name information included in the first map image (steps S301 and S302). The N value “2” of the record with ID “1” in the score information of FIG. 13 is accumulated in this way. The score information in FIG. 13 is stored in a recording medium (not shown), and the ID is an identifier for identifying a record and corresponds to the ID in FIG. The ID, the number N, the region consistency R, the spatial consistency S1, the spatial consistency S2, the text consistency T1, the text consistency T2, and the score are associated with each other. The number specifying unit 16 or the like adds information to the score information every time the number N or the like is acquired, and finally the output unit 22 calculates the score using the information and adds it to the score information.

  Next, the region matching degree calculation unit 17 determines that two or more place name information has been received (step S303), and stores the latitude / longitude corresponding to the received place name information “Himeji” and “Akashi” in the place name database. And obtain the region MBR ({Himeji, Akashi}) using them. Specifically, as shown in FIG. 14, a region MBR ({Himeji, Akashi}) having a diagonal between Himeji and Akashi is specified. Further, the region matching degree calculation unit 17 uses the place name character strings “Kobe”, “Akashi”, and “Himeji” included in the first map image “ABC-MAP.gif”, and the map image “ABC-MAP.gif”. EBR ({Kobe, Akashi, Himeji}, ABC-MAP.gif). Specifically, as shown in FIG. 14, the region matching degree calculation unit 17 specifies the region MBR ({Kobe, Akashi, Himeji}) in the same manner as the region MBR ({Himeji, Akashi}). Further, as shown in FIG. 15, the region matching degree calculation unit 17 specifies a region MBR ({Kobe, Akashi, Himeji}) in the first map image “ABC-MAP.gif”. At that time, the position (x11, y11) of the character string information shown in FIG. 12 is used as the coordinates of the place name character string “Kobe”. After that, the region matching degree calculation unit 17 performs the same processing as in FIG. 14 on expanding the region MBR ({Kobe, Akashi, Himeji}) in the map image to the entire map image as indicated by the arrows in FIG. An area EBR ({Kobe, Akashi, Himeji}, ABC-MAP.gif) obtained by expanding the area MBR ({Kobe, Akashi, Himeji}) is specified. Then, the region matching degree calculation unit 17 calculates the region matching degree R using the above formula. Here, it is assumed that the region matching degree R is “0.32.” Then, as shown in FIG. 13, the region matching degree calculation unit 17 accumulates the region matching degree R “0.32” in the record of the score information ID “1” (step S304).

  Next, the actual map information acquisition unit 18 acquires actual map information including the place name character strings “Kobe”, “Akashi”, and “Himeji” included in the first map image, and records the acquired actual map information not shown. Accumulate on the medium (step S305). The actual measurement map information is assumed to be the same as the actual measurement map information shown in FIG.

  Next, the space matching degree calculation unit 19 sets the place name character string “Akashi” closest to the center of the image in the first map image as the reference place name. Then, the space matching degree calculation unit 19 calculates the distance between the place name character string “Akashi” and the place name character strings “Himeji” and “Kobe”, and obtains the order of the distances. In this case, it is assumed that (Distance between Akashi and Kobe)> (Distance between Akashi and Himeji), (Distance between Akashi and Kobe) is No. 1, and (Distance between Akashi and Himeji) is No. 2. In addition, the position (x11, y11) of Kobe, Akashi, Himeji, etc. are used for calculation of the distance.

  Further, the spatial matching degree calculation unit 19 also calculates the distance to Kobe and the distance to Himeji based on the position of Akashi in the actual map information acquired by the actual map information acquisition unit 18. Ask for order. In this case, it is assumed that (Distance between Akashi and Himeji)> (Distance between Akashi and Kobe), (Distance between Akashi and Himeji) is No. 1, and (Distance between Akashi and Himeji) is No. 2. In calculating the distance, the latitude and longitude of Kobe, Akashi, and Himeji are used.

  After that, the spatial consistency calculation unit 19 calculates the spatial consistency S1 as described above using Spearman's rank correlation coefficient. In this case, since the order of the map image and the actually measured map information is completely reversed, S1 = 0. Then, as shown in FIG. 13, the spatial matching degree calculation unit 19 accumulates the spatial matching degree S1 “0” in the record of ID “1” of the score information (step S306).

  Next, in the first map image, the spatial matching degree calculation unit 19 sets the place name character strings “Himeji” and “Kobe” located substantially opposite to the image center as the first and second place name character strings. . Then, in the first map image, the spatial matching degree calculation unit 19 determines whether the position corresponding to “Akashi” is on the right side or the left side with respect to the straight line passing through the positions corresponding to the place name character strings “Himeji” and “Kobe”. Judging. The right and left sides here are assumed to be the right and left when looking toward Himeji from Kobe. Therefore, in the map image of FIG. 8, Akashi is on the left side.

  In addition, the spatial matching degree calculation unit 19 has a position corresponding to “Akashi” with respect to a straight line passing through positions corresponding to “Himeji” and “Kobe” in the actual measurement map information acquired by the actual measurement map information acquisition unit 18. Determine whether it is right or left. In this case as well, the right and left sides are the right and left when looking toward Himeji from Kobe. Therefore, as is clear from FIG. 14, Akashi is on the left side.

  Therefore, c (Himeji, Kobe, Akashi) = 1. In this case, since k = 1, the space matching degree calculation unit 19 calculates the space matching degree S2 “1”. Then, as shown in FIG. 13, the spatial consistency degree calculation unit 19 accumulates the spatial consistency degree S2 “1” in the record of ID “1” of the score information (step S307).

  Next, the surrounding text acquisition unit 20 acquires a URL corresponding to the first map image “ABC-MAP.gif” in FIG. 12, and accesses the URL. Then, the surrounding text acquisition unit 20 acquires the surrounding text of the first map image “ABC-MAP.gif” in the accessed page, and stores it in a recording medium (not shown) (step S308).

  Next, the text matching degree calculation unit 21 performs a morphological analysis on the surrounding text acquired by the surrounding text acquisition unit 20, acquires a proper noun that is a place name, and stores it in a recording medium (not shown). In this case, it is assumed that “Akashi”, “Himeji”, “Kobe”, “Sannomiya”, “Nishi-Akashi”, and “Shin-Osaka” have been acquired. In addition, the text matching degree calculation unit 21 counts the number of place names that are common to the place character string included in the first map image and the place names included in the surrounding text. In this case, since “Akashi”, “Himeji” and “Kobe” are common, the number is “3”. Further, the text matching degree calculation unit 21 counts the number of place names included in the surrounding text. In this case, the number is “6”. Then, the text matching level calculation unit 21 calculates the text matching level T1 “0.5” (≈3 / 6) by using the above-described first formula. Then, as shown in FIG. 13, the text consistency degree calculation unit 21 accumulates the text consistency degree T1 “0.5” in the record of the score information ID “1” (step S309).

  Next, the text matching degree calculation unit 21 specifies an area MBR ({Akashi, Himeji, Kobe, Sannomiya, Nishi-Akashi, Shin-Osaka}) in the measured map information corresponding to the place name included in the surrounding text. In addition, the text matching degree calculation unit 21 specifies the area EBR ({Kobe, Akashi, Himeji}, ABC-MAP.gif) of the actually measured map information corresponding to the first map image. Then, it is assumed that the text matching degree T2 calculated by using the above-mentioned second formula is “0.64”. Then, as shown in FIG. 13, the text matching degree calculation unit 21 accumulates the text matching degree T2 “0.64” in the record of ID “1” of the score information (step S310).

Next, the output unit 22 calculates a score using the number N or the like. Here, it is assumed that the score formula is as follows.
Score = N + R + S1 + S2 + T1 + T2
Then, the output unit 22 calculates the score “4.46” of the first map image, and accumulates the score in the record with ID “1” as shown in FIG. 13 (step S311).

  Thereafter, the number specifying unit 16 and the like repeatedly execute processing such as specifying the number N for the second and subsequent map images (steps S302 to S313). As a result, the score information is as shown in FIG.

  And the output part 22 produces | generates the screen which displays the map image corresponding to a score on a display so that it may become descending order of the score information of FIG. 13, and displays it as shown in FIG. 16 (step S105). In this specific example, the map image “XYZ-MAP.jpg” has the highest score, and the map image “ABC-MAP.gif” has the second highest score. By viewing this screen, the user can browse images corresponding to the input place name information “Himeji” and “Akashi” in the order of ranking.

  As described above, according to the map image acquisition device 1 according to the present embodiment, a map image corresponding to one or more place name information received by the place name information receiving unit 11 can be acquired. Even if an image search is performed using the place name information as a search key, it is not always possible to acquire a map image. However, in the map image acquisition apparatus 1, determination by the place name determination unit 13 or determination by the threshold value determination unit 14 is performed. By performing the above, it is possible to acquire a map image that is highly likely to be a map image. In addition, even if there is a public image that does not correspond to a change of a place name in the real world, a facility closure, or the like, that is, a public image that includes a place name that does not currently exist, the threshold determination unit 14 By using the latest place name database to determine place names, place name strings that do not currently exist can be determined not to be place name strings, and a highly accurate map image corresponding to the real world. Can also be possible. In addition, since the obtained map images are ranked and output using the number N, the region matching degree R, the spatial matching degrees S1, S2, and the text matching degrees T1, T2, a more appropriate map image is output to the upper level. It becomes possible to improve the convenience for the user.

  In the present embodiment, the case where the spatial matching degree calculation unit 19 calculates the spatial matching degree using the actual measurement map information acquired by the actual measurement map information acquisition unit 18 is described, but this need not be the case. The space matching degree calculation unit 19 can acquire the position on the map based on the actual measurement corresponding to the place name character string using, for example, a place name database. Therefore, the spatial consistency degree calculation unit 19 may calculate the spatial consistency degrees S1 and S2 using the position (for example, latitude / longitude) acquired using the place name database. Even in that case, the spatial matching degree calculation unit 19 can calculate the spatial matching degree indicating the degree of matching between the map image and the map based on the actual measurement. In the description of the processing for calculating the degree of spatial matching described above, it is obvious that only the position in the actually measured map information is used and no figure is used. Note that when the degree of spatial matching is calculated without using the actual map information, the map image acquisition apparatus 1 may not include the actual map information acquisition unit 18.

Further, in the present embodiment, the case where the text matching level calculation unit 21 calculates the text matching level T2 has been described, but this need not be the case. In that case, the output unit 22 may not perform ranking of the map image using the text matching degree T2.
In the present embodiment, the case where the text matching level calculation unit 21 calculates the text matching level T1 has been described. However, this need not be the case. In that case, the output unit 22 may not perform ranking of the map image using the text matching degree T1.

In the present embodiment, the text matching degree calculation unit 21 may calculate a text matching degree indicating the degree of matching regarding the map image and the place name included in the surrounding text other than the text matching degrees T1 and T2. Needless to say, it is good.
In the present embodiment, the text matching degree need not be calculated. In that case, the map image acquisition device 1 may not include the surrounding text acquisition unit 20 and the text matching degree calculation unit 21.

In the present embodiment, the case where the space matching degree calculation unit 19 calculates the space matching degree S2 has been described, but this need not be the case. In that case, the output unit 22 may not perform ranking of map images using the spatial matching degree S2.
In the present embodiment, the case where the spatial matching degree calculation unit 19 calculates the spatial matching degree S1 has been described, but this need not be the case. In that case, the output unit 22 may not perform ranking of map images using the spatial matching degree S1.

In the present embodiment, the spatial matching degree calculation unit 19 may calculate a spatial matching degree indicating the degree of matching between the map image and the map based on the actual measurement other than the spatial matching degrees S1 and S2. Needless to say.
In the present embodiment, the degree of spatial matching need not be calculated. In that case, the map image acquisition device 1 may not include the actual measurement map information acquisition unit 18 and the spatial matching degree calculation unit 19.

  In the present embodiment, the case where the region matching degree calculation unit 17 calculates the region matching degree R has been described, but this need not be the case. In that case, the output unit 22 may not perform ranking of the map image using the region matching degree R. In that case, the map image acquisition apparatus 1 may not include the region matching degree calculation unit 17.

  In the present embodiment, the case where the number specifying unit 16 specifies the number N of place name information included in the map image has been described, but this need not be the case. In that case, the output unit 22 may not perform ranking of map images using the number N. In this case, the map image acquisition device 1 may not include the number specifying unit 16.

  Moreover, although this Embodiment demonstrated the case where the map image containing more place names than a threshold value was acquired, it may not be so. In that case, the map image acquisition apparatus 1 may not include the threshold value determination unit 14. In that case, the map image acquisition unit 15 may acquire a map image that is a public image determined to include the place name information by the place name determination unit 13.

  In the present embodiment, the case where the place name determination unit 13 determines whether or not the received place name information is included has been described. In the case where the determination by the place name determination unit 13 is not performed, for example, the map image acquisition unit 15 determines whether the place name information is included in the public image without performing character recognition, and according to the determination result. You may acquire a map image. In that case, for example, a place name (a place name described by a character code) corresponding to a place name (a place name of the image) included in the image may be included in the file name or header of the image. And map image acquisition part 15 may judge whether a public image is a map image containing place name information using information, such as the file name. In addition, when the determination by the place name determination unit 13 is not performed, the map image acquisition device 1 may not include the character recognition unit 12 or the place name determination unit 13. Note that when the result of character recognition is used in subsequent processing such as the number identification unit 16, the map image acquisition device 1 includes the character recognition unit 12 even when the place name determination unit 13 does not perform the determination. It may be.

  In the present embodiment, the map image acquisition device 1 has been described. However, the device for ranking the map image includes the spatial matching degree calculation unit 19, the surrounding text acquisition unit 20, the text matching degree calculation unit 21, the output unit 22, and the like. It may be used as In that case, for example, the map image ranking device includes: a map image storage unit that stores a plurality of map images; a calculation unit that calculates a ranking index (score) for ranking for each map image; An output unit that outputs a map image so as to be ranked higher as the calculated ranking index is higher may be provided. Note that the calculation unit may have the same function as any one or more of the spatial matching level calculation unit 19 and the text matching level calculation unit 21. When the calculation unit has the same function as the text matching degree calculation unit 21, the map image ranking device may include the surrounding text acquisition unit 20. When acquiring the surrounding text, for example, in the map image storage unit, information specifying the page from which the map image is acquired in association with the map image may be stored. The peripheral text itself corresponding to the map image may be stored. In the latter case, the surrounding text acquisition unit 20 can acquire the surrounding text from the map image storage unit. In addition, for example, the output unit outputs a map image so as to rank higher as the degree of spatial consistency is higher, or outputs a map image so as to rank higher as the degree of text consistency is higher, This is the same as the present embodiment.

  In the above embodiment, the map image acquisition device 1 is a client as shown in FIG. 5, but the map image acquisition device 1 itself may be a server device in a server / client system. . In that case, the output unit or the reception unit may receive an input or output information through a communication line.

  In the above embodiment, each process or each function may be realized by centralized processing by a single device or a single system, or may be distributedly processed by a plurality of devices or a plurality of systems. It may be realized by doing.

  In the above embodiment, information related to processing executed by each component, for example, information received, acquired, selected, generated, transmitted, or received by each component In addition, information such as threshold values, mathematical formulas, addresses, etc. used by each component in processing is retained temporarily or over a long period of time on a recording medium (not shown) even when not explicitly stated in the above description. It may be. Further, the storage of information in the recording medium (not shown) may be performed by each component or a storage unit (not shown). Further, reading of information from the recording medium (not shown) may be performed by each component or a reading unit (not shown).

  In the above embodiment, when information used by each component, for example, information such as a threshold value, an address, and various setting values used by each component may be changed by the user Even if it is not specified in the above description, the user may be able to change the information as appropriate, or it may not be. If the information can be changed by the user, the change is realized by, for example, a not-shown receiving unit that receives a change instruction from the user and a changing unit (not shown) that changes the information in accordance with the change instruction. May be. The change instruction received by the receiving unit (not shown) may be received from an input device, information received via a communication line, or information read from a predetermined recording medium, for example. .

  In the above embodiment, when two or more components included in the map image acquisition apparatus 1 have communication devices, input devices, etc., the two or more components have a physically single device. Or may have separate devices.

  In the above embodiment, each component may be configured by dedicated hardware, or a component that can be realized by software may be realized by executing a program. For example, each component can be realized by a program execution unit such as a CPU reading and executing a software program recorded on a recording medium such as a hard disk or a semiconductor memory. In addition, the software which implement | achieves the map image acquisition apparatus 1 in the said embodiment is the following programs. In other words, this program publishes an image of a map image, which is an image including place name information received by the place name information receiving section and the place name information receiving section, which receives one or more place name information as place name information. It is a program for functioning as an output part which outputs the map image acquisition part acquired from the server which is, and the map image which the map image acquisition part acquired.

  In the program, the functions realized by the program do not include functions that can be realized only by hardware. For example, functions that can be realized only by hardware such as a modem or an interface card in an acquisition unit that acquires information, an output unit that outputs information, and the like are not included in at least the functions realized by the program.

  Further, this program may be executed by being downloaded from a server or the like, and a program recorded on a predetermined recording medium (for example, an optical disk such as a CD-ROM, a magnetic disk, a semiconductor memory, or the like) is read out. May be executed by Further, this program may be used as a program constituting a program product.

  Further, the computer that executes this program may be singular or plural. That is, centralized processing may be performed, or distributed processing may be performed.

  FIG. 17 is a schematic diagram illustrating an example of an external appearance of a computer that executes the program and realizes the map image acquisition apparatus 1 according to the embodiment. The above-described embodiment can be realized by computer hardware and a computer program executed on the computer hardware.

  In FIG. 17, a computer system 900 includes a computer 901 including a CD-ROM (Compact Disk Read Only Memory) drive 905, an FD (Floppy (registered trademark) Disk) drive 906, a keyboard 902, a mouse 903, a monitor 904, and the like. Is provided.

  FIG. 18 is a diagram showing an internal configuration of the computer system 900. In FIG. 18, in addition to the CD-ROM drive 905 and the FD drive 906, a computer 901 is connected to an MPU (Micro Processing Unit) 911, a ROM 912 for storing a program such as a bootup program, and the MPU 911. A RAM (Random Access Memory) 913 that temporarily stores program instructions and provides a temporary storage space, a hard disk 914 that stores application programs, system programs, and data, and an MPU 911 and a ROM 912 are interconnected. And a bus 915. The computer 901 may include a network card (not shown) that provides connection to the LAN.

  A program that causes the computer system 900 to execute the function of the map image acquisition apparatus 1 according to the above-described embodiment is stored in the CD-ROM 921 or the FD 922, inserted into the CD-ROM drive 905 or the FD drive 906, and the hard disk 914. May be forwarded to. Instead, the program may be transmitted to the computer 901 via a network (not shown) and stored in the hard disk 914. The program is loaded into the RAM 913 when executed. The program may be loaded directly from the CD-ROM 921, the FD 922, or the network.

  The program does not necessarily include an operating system (OS) or a third-party program that causes the computer 901 to execute the functions of the map image acquisition device 1 according to the above-described embodiment. The program may include only a part of an instruction that calls an appropriate function (module) in a controlled manner and obtains a desired result. How the computer system 900 operates is well known and will not be described in detail.

  Further, the present invention is not limited to the above-described embodiment, and various modifications are possible, and it goes without saying that these are also included in the scope of the present invention.

  As mentioned above, according to the map image acquisition apparatus etc. by this invention, the effect that the map image corresponding to one or more place name information can be acquired is acquired, and it is useful as an apparatus etc. which acquire a map image.

DESCRIPTION OF SYMBOLS 1 Map image acquisition apparatus 11 Place name information reception part 12 Character recognition part 13 Place name determination part 14 Value determination part 15 Map image acquisition part 16 Number specific | specification part 17 Area | region matching degree calculation part 18 Actual measurement map information acquisition part 19 Spatial consistency degree calculation part 20 Peripheral text acquisition unit 21 Text matching degree calculation unit 22 Output unit

Claims (11)

A place name information receiving unit for receiving one or more pieces of place name information which are information of place names;
A character recognition unit that performs character recognition on a public image that is a public image on a server;
A place name determining unit that determines whether or not the place name information received by the place name information receiving unit is included in a character string that has been subjected to character recognition by the character recognition unit, included in the public image;
It is determined by accessing a place name database that associates place name information with coordinates corresponding to the place name information, whether the character string that has been subjected to character recognition by the character recognition unit included in the public image is a place name. A threshold determination unit that determines whether or not the public image includes more place names than the threshold;
A map that is a public image that is determined to include the place name information received by the place name information receiving unit by the place name determining unit and that is determined to include more place names than the threshold value by the threshold value determining unit. A map image acquisition unit for acquiring images;
An output unit that outputs a map image acquired by the map image acquisition unit. A number specifying unit for specifying the number of one or more place name information received by the place name information receiving unit included in the map image acquired by the map image acquiring unit;
And the output unit, the number is specified unit outputs the map image to ranked high larger the map image number identified, the map image acquisition apparatus according to claim 1. The place name information receiving unit receives a plurality of place name information,
A map based on actual measurement corresponding to a minimum area in a map based on actual measurement, including points corresponding to a plurality of place name information attached by the place name information receiving unit, and a map image acquired by the map image acquisition unit A region matching degree calculation unit that calculates a region matching degree according to the degree of overlap with the region of
The output unit outputs the map image to ranked high higher the area matching degree, the map image acquisition apparatus according to claim 1 or claim 2, wherein. A map degree acquired by the map image acquisition unit, and a spatial matching degree calculation unit that calculates a spatial matching degree indicating a degree of matching between the map based on the actual measurement, and
The output unit outputs the map image to ranked high higher the spatial matching degree, the map image acquisition apparatus according to any one of claims 1 to 3. The spatial matching degree calculation unit is based on an order of distances from a point corresponding to one place name character string to a point corresponding to another place name character string in the map image acquired by the map image acquisition unit, and based on an actual measurement. The map image acquisition apparatus of Claim 4 which calculates the spatial consistency degree which shows the correlation with the order of the distance from the point corresponding to said one place name character string to the point corresponding to another place name character string in a map. The spatial matching degree calculation unit includes a position of a point corresponding to another place name character string with respect to a straight line connecting points corresponding to the first and second place name character strings in the map image acquired by the map image acquisition unit; Calculating a spatial matching degree indicating a degree of coincidence with a position of a point corresponding to another place name character string with respect to a straight line connecting points corresponding to the first and second place name character strings in a map based on actual measurement; The map image acquisition device according to claim 4 or 5 . From a page including the map image acquired by the map image acquisition unit, a peripheral text acquisition unit that acquires peripheral text that is text around the map image;
A text matching degree calculating unit that calculates a text matching degree indicating a degree of matching with respect to a map image acquired by the map image acquiring unit and a place name included in the surrounding text acquired by the surrounding text acquiring unit;
And the output unit, the output of the map image so that the text matching degree is ranked high higher, the map image acquisition apparatus according to any one of claims 1 to 6. The text matching degree calculation unit is determined to be a place name using a place name database that associates place name information and coordinates corresponding to the place name information among character strings included in the map image acquired by the map image acquisition unit. The map image acquisition apparatus according to claim 7 , wherein a text matching degree is calculated in accordance with a degree of matching between a place name character string that is a character string and a place name included in the surrounding text. The text matching degree calculation unit includes a point corresponding to a place name included in the surrounding text, a minimum region in a map based on actual measurement, and an actual measurement corresponding to the map image acquired by the map image acquisition unit. The map image acquisition device according to claim 7 or 8 , wherein a text matching degree is calculated according to a degree of overlap with an area of a map based on the map. A map image acquisition method processed using a place name information reception unit, a character recognition unit, a place name determination unit, a threshold value determination unit, a map image acquisition unit, and an output unit,
A place name information receiving step in which the place name information receiving unit receives one or more pieces of place name information that are place name information;
A character recognition step in which the character recognition unit performs character recognition on a public image that is an image publicized on a server;
A place name determining step for determining whether the place name information received in the place name information receiving step is included in the character string that has been subjected to character recognition in the character recognition step, included in the public image;
A place name that associates place name information with coordinates corresponding to the place name information, whether or not the character string that has been character-recognized in the character recognition step included in the public image is a place name. A threshold determination step for determining by accessing the database and determining whether the public image includes more place names than the threshold;
The map image acquisition unit determines that the place name information received in the place name information reception step is included in the place name determination step, and determines that more place names than the threshold value are included in the threshold value determination step. A map image acquisition step of acquiring a map image which is a published image ,
An output step in which the output unit outputs the map image acquired in the map image acquisition step. Computer
A place name information receiving unit for receiving one or more place name information, which are place name information,
A character recognition unit that performs character recognition on a public image that is a public image on the server,
A place name determining unit that determines whether or not the place name information received by the place name information receiving unit is included in a character string that has been subjected to character recognition by the character recognition unit, included in the public image;
It is determined by accessing a place name database that associates place name information with coordinates corresponding to the place name information, whether the character string that has been subjected to character recognition by the character recognition unit included in the public image is a place name. , A threshold determination unit that determines whether the public image includes more place names than the threshold,
A map that is a public image that is determined to include the place name information received by the place name information receiving unit by the place name determining unit and that is determined to include more place names than the threshold value by the threshold value determining unit. A map image acquisition unit for acquiring images,
The program for functioning as an output part which outputs the map image which the said map image acquisition part acquired.

Images