Map expression data processing device, method for generating corresponding information, and program

Abstract

In the past, it has not been possible to automatically acquire coordinate information and position information of place names and the like on map expression data in a corresponding and associated manner. The present invention enables automatic acquisition of coordinate information and position information of place names and the like on map expression data in a corresponding and associated manner by a map expression data processing device that includes a map expression data reception unit that receives map expression data, a character string acquisition unit that acquires a character string from the map expression data, a coordinate information acquisition unit that acquires coordinate information corresponding to the acquired character string, a position information acquisition unit that acquires position information corresponding to place information that is a character string using a place dictionary having one or more pieces of place position information that correspond and are associated with position information, and a corresponding output unit that outputs the coordinate information and the position information in a corresponding and associated manner.

Description

TECHNICAL FIELD

[0001] The present application relates to a map expression data processing apparatus and the like that automatically acquires and outputs correspondence information indicating correspondence of a place name and position information acquired from map expression data. BACKGROUND

[0002] Conventionally, there is a map information system that enables a user of one or more first terminal apparatuses to register a hand-drawn map or the like and a map illustration and the like, which are made or prepared by the user himself / herself, and enables a user of one or more second terminal apparatuses to use the map and to assign geographical information as information of an object such as a landmark to the map (see Patent Literature 1).

[0003] PRIOR ART DOCUMENTS

[0004] PATENT LITERATURE

[0005] Patent Literature 1: Japanese Patent Application Publication No. 2011-158518 SUMMARY

[0006] PROBLEMS TO BE SOLVED BY THE INVENTION

[0007] However, in the conventional technique, it is not possible to automatically acquire coordinate information of a place name and the like on map expression data of a hand-drawn map and a map illustration and the like in correspondence with position information.

[0008] SOLUTION TO PROBLEM

[0009] The map expression data processing apparatus of the first application includes a place dictionary storage section that stores one or more pieces of place position information that are information that associates place information of a determined place with position information of a determined position in correspondence with each other; a map expression data reception section that receives map expression data of a map of a kind that is any one or more of a map illustration, a hand-drawn map, and a sketch map; a character string acquisition section that acquires a character string from the map expression data; a coordinate information acquisition section that acquires coordinate information that is information that determines a coordinate position corresponding to the character string acquired by the character string acquisition section and that is information that determines a relative coordinate position in the map expression data; a position information acquisition section that acquires position information that corresponds to the place information of the character string acquired by the character string acquisition section using the place dictionary; and a correspondence output section that outputs the coordinate information acquired by the coordinate information acquisition section in correspondence with the position information acquired by the position information acquisition section.

[0010] According to such a configuration, it is possible to automatically acquire coordinate information of a place name and the like on map expression data in correspondence with position information.

[0011] Further, in the map expression data processing device according to the second aspect of the present invention, the string acquisition section acquires two or more strings from the map expression data, and the classification section determines whether each of the two or more strings acquired by the string acquisition section is a string of a certain place, and the corresponding output section outputs coordinate information and position information corresponding to the string determined by the classification section as the string of the certain place in association with each other.

[0012] According to such a configuration, a proper string of a place name or the like on the map expression data can be extracted, and coordinate information and position information of the place name or the like can be automatically acquired in association with each other with high precision.

[0013] Further, in the map expression data processing device according to the third aspect of the present invention, the learner storage section gives two or more pieces of learning source information including a string having two or more characters and a label related to the number of occurrences in the place dictionary and being any one of two or more labels to a learning module of machine learning, executes the learning module, and stores the acquired learner, the classification section classifies each of the two or more strings acquired by the string acquisition section into one of place information and non-place information by a technique of machine learning using the learner, and the corresponding output section outputs coordinate information and position information corresponding to the string classified as the place information by the classification section in association with each other.

[0014] According to such a configuration, a proper string of a place name or the like on the map expression data can be extracted with high precision, and coordinate information and position information of the place name or the like can be automatically acquired in association with each other.

[0015] Further, in the map expression data processing device according to any one of the first to third aspects of the present invention, the feature place detection section detects a feature place that is a place represented by a feature pattern that is a predetermined characteristic pattern from the map expression data, the coordinate information acquisition section acquires coordinate information corresponding to the feature pattern, and the position information acquisition section acquires position information corresponding to place information that is a string corresponding to the feature pattern using the place dictionary.

[0016] According to such a configuration, a feature point of a pattern on the map expression data can be extracted, and coordinate information and position information of the feature point of the pattern can be automatically acquired in association with each other.

[0017] Further, in the map expression data processing device according to the fifth aspect of the present invention, the feature pattern is one or more patterns of an intersection or a bridge.

[0018] According to such a configuration, a location of an intersection or a bridge as a graphic feature point on the map expression data can be acquired, and coordinate information of the location is automatically acquired in association with the position information.

[0019] Further, with respect to any one of the first to fifth inventions, the string acquisition section of the map expression data processing apparatus of the sixth invention acquires three or more strings from the map expression data, and further includes a grouping section that determines two or more pieces of position information that indicate locations close to satisfying a predetermined condition from among the three or more pieces of position information acquired by the position information acquisition section, and the corresponding output section outputs each of the two or more pieces of position information determined by the grouping section in association with each of the two or more pieces of coordinate information acquired by the coordinate information acquisition section.

[0020] According to such a configuration, a proper string of a place name or the like on the map expression data can be extracted, and coordinate information of the place name or the like is automatically acquired in association with the position information.

[0021] Further, with respect to the sixth invention, the map expression data processing apparatus of the seventh invention further includes an accurate map information storage section that stores accurate map information as accurate map information, a region determination information acquisition section that acquires region determination information that determines a region in the accurate map information that contains a region indicated by the two or more pieces of position information determined by the grouping section, and an accurate map information output section that outputs the accurate map information in a manner that makes it possible to recognize the region determined by the region determination information.

[0022] According to such a configuration, the range of the accurate map information corresponding to the map expression data can be made explicit.

[0023] Further, with respect to the seventh invention, the map expression data processing apparatus of the eighth invention has a region determination information acquisition section that includes: a first region determination information acquisition unit that acquires first region determination information that determines a region in accurate map information that is a region including positions indicated by two or more pieces of position information determined by the grouping section; an outside string determination unit that determines a string corresponding to each of one or more pieces of position information located at the outermost side among the pieces of position information acquired by the position information acquisition section; a size information acquisition unit that acquires size information related to a size from the outside of the one or more strings determined by the outside string determination unit to a corner of the map expression data; a distance information acquisition unit that acquires distance information that determines a distance corresponding to the size information acquired by the size information acquisition unit using two or more groups of the coordinate information acquired by the coordinate information acquisition section and the position information acquired by the position information acquisition section; and a second region determination information acquisition unit that acquires second region determination information that determines a second region that is a region expanded from the first region determined by the first region determination information using the distance information, and the accurate map information output section outputs the accurate map information in a manner that knows the region determined by the second region determination information.

[0024] According to such a configuration, it is possible to specify an appropriate range of accurate map information corresponding to the map expression data.

[0025] Further, with respect to the seventh or eighth invention, the map expression data processing apparatus of the ninth invention includes information of two points that constitute a rectangle in the region determination information or the second region determination information, and further includes: an orientation information acquisition section that acquires orientation information related to an orientation of the map expression data using the region determination information or the second region determination information; and an orientation information output section that outputs the orientation information.

[0026] According to such a configuration, it is possible to acquire orientation information related to an orientation of the map expression data.

[0027] Further, with respect to any one of the first to fourth inventions, the map expression data processing apparatus of the tenth invention further includes: a relationship information acquisition section that acquires relationship information related to a relationship between correspondence information that is a group of coordinate information and position information corresponding to three or more strings acquired by the string acquisition section; a correspondence information acquisition section that acquires only two or more pieces of correspondence information corresponding to relationship information that satisfies a relationship decided in advance among the three or more pieces of correspondence information using the relationship information; and a correspondence output section that stores the two or more pieces of correspondence information acquired by the correspondence information acquisition section.

[0028] According to such a configuration, it is possible to automatically acquire appropriate correspondence information that associates coordinate information and position information of a place name or the like on the map expression data.

[0029] Further, with respect to the tenth invention, the map expression data processing apparatus of the eleventh invention has a relationship information acquisition section that acquires relative relationship information between corresponding information corresponding to two respective strings among the three or more strings acquired by the string acquisition section, in units of groups of two strings. The corresponding information acquisition section has a determination unit that groups the three or more pieces of relationship information acquired by the relationship information acquisition section and determines relationship information that does not enter the group, and a corresponding information acquisition unit that excludes corresponding information corresponding to the relationship information that does not enter the group from the corresponding information corresponding to the three or more respective strings acquired by the string acquisition section and acquires only the remaining two or more pieces of corresponding information.

[0030] According to such a configuration, appropriate corresponding information that associates coordinate information of a place name or the like on map expression data with position information can be automatically acquired.

[0031] Further, with respect to the tenth or eleventh invention, the map expression data processing apparatus of the twelfth invention has a relationship information acquisition section that acquires distance relationship information that is relationship information indicating a relationship between a relative distance that is a difference in coordinate information of two strings and an absolute distance that is a difference in position information of the two strings, in units of groups of two strings. The corresponding information acquisition section determines whether three or more respective pieces of distance relationship information satisfy a predetermined relationship, excludes corresponding information corresponding to distance relationship information that does not satisfy the predetermined relationship from the corresponding information corresponding to the three or more respective strings acquired by the string acquisition section, and acquires only the remaining two or more pieces of corresponding information.

[0032] According to such a configuration, appropriate corresponding information that associates coordinate information of a place name or the like on map expression data with position information can be automatically acquired.

[0033] Further, with respect to the twelfth invention, the map expression data processing apparatus of the thirteenth invention further has a scale acquisition section that acquires a representative value of distance relationship information from the three or more pieces of distance relationship information and acquires scale information of the map expression data using the representative value of distance relationship information, and a scale information output section that outputs the scale information.

[0034] According to such a configuration, appropriate corresponding information that associates coordinate information of a place name or the like on map expression data with position information can be automatically acquired.

[0035] Further, with respect to the fourteenth application, the map expression data processing apparatus of the fifteenth application further comprises: an additional place acquisition section that acquires place information of the range indicated by the area determination information from a place dictionary; an additional position acquisition section that acquires position information corresponding to the place information from the place dictionary; an additional coordinate acquisition section that acquires coordinate information corresponding to the position information; and a place registration section that registers the place information at a position on the map expression data and at a position indicated by the coordinate information.

[0036] According to such a configuration, it is possible to acquire a range in the real world supported by the map expression data.

[0037] Further, with respect to the fourteenth application, the map expression data processing apparatus of the fifteenth application further comprises: an additional place acquisition section that acquires place information of the range indicated by the area determination information from a place dictionary; an additional position acquisition section that acquires position information corresponding to the place information from the place dictionary; an additional coordinate acquisition section that acquires coordinate information corresponding to the position information; and a place registration section that registers the place information at a position on the map expression data and at a position indicated by the coordinate information.

[0038] According to such a configuration, it is possible to add new place information to the map expression data.

[0039] Further, with respect to the tenth or eleventh application, the map expression data processing apparatus of the sixteenth application further comprises: a relationship information acquisition section that acquires, as relationship information indicating a relationship between a relative angle that is an angle obtained from coordinate information of two strings and an absolute angle that is an angle obtained from position information of the two strings, angle relationship information for each of two or more groups of the two strings; a corresponding information acquisition section that determines whether each of the three or more pieces of angle relationship information satisfies a predetermined relationship, excludes, from corresponding information corresponding to each of the three or more strings acquired by the string acquisition section, corresponding information corresponding to angle relationship information that does not satisfy the predetermined relationship, and acquires only the remaining two or more pieces of corresponding information.

[0040] According to such a configuration, it is possible to automatically acquire appropriate corresponding information that associates coordinate information of a place name or the like on the map expression data and position information.

[0041] Further, with respect to the sixteenth application, the map expression data processing apparatus of the seventeenth application further comprises: a direction information acquisition section that acquires a representative value of the angle relationship information from the three or more pieces of angle relationship information, and acquires direction information indicating a direction of the map expression data using the representative value of the angle relationship information; and a direction information output section that outputs the direction information.

[0042] According to such a configuration, it is possible to acquire direction information of the map expression data.

[0043] Further, the map representation data processing apparatus of the eighteenth aspect includes a map representation data storage section that stores map representation data; a correspondence information storage section that stores two or more pieces of correspondence information that are a group of coordinate information output by the map representation data processing apparatus of any one of the first through seventeenth aspects and position information acquired by the position information acquisition section; a current position information acquisition section that acquires current position information that specifies a current position; a coordinate information acquisition section that acquires coordinate information corresponding to the current position information using the two or more pieces of correspondence information; a data construction section that constructs current position additional map representation data that is current position additional map representation data of a position indicated by the explicit coordinate information on the map representation data; and a map representation data output section that outputs the current position additional map representation data.

[0044] According to such a configuration, the current position can be indicated on the map representation data using the two or more pieces of correspondence information that are automatically acquired.

[0045] Effects of the Invention

[0046] According to the map representation data processing apparatus based on the present invention, the coordinate information of a place name or the like on the map representation data and the position information can be automatically acquired in association with each other. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 is a block diagram of a learning device A in the first embodiment.

[0048] Figure 2 is a block diagram of a classification device B.

[0049] Figure 3 is a flowchart illustrating an example of the operation of the learning device A.

[0050] Figure 4 is a flowchart illustrating an example of the operation of the classification device B.

[0051] Figure 5 is a diagram illustrating a specific example of the operation of the learning device A and the classification device B.

[0052] Figure 6 is a diagram showing an example of the learning source information.

[0053] Figure 7 is a block diagram of a map representation data processing apparatus C in the second embodiment.

[0054] Figure 8 is a flowchart illustrating an example of the operation of the map representation data processing apparatus C.

[0055] Figure 9 is a flowchart illustrating an example of the grouping processing.

[0056] Figure 10 is a flowchart illustrating an example of the region determination processing.

[0057] Figure 11 is a flowchart illustrating an example of the orientation acquisition processing.

[0058] Figure 12 is a flowchart illustrating an example of the output processing.

[0059] Figure 13 is a diagram showing an example of the map expression data.

[0060] Figure 14 is a diagram showing an example of the map expression data.

[0061] Figure 15 is a diagram showing an example of the place dictionary.

[0062] Figure 16 is a diagram showing an example of the correspondence information management table.

[0063] Figure 17 is a block diagram of the map expression data processing apparatus D in the third embodiment.

[0064] Figure 18 is a flowchart illustrating an example of the operation of the map expression data processing apparatus D.

[0065] Figure 19 is a block diagram of the map expression data processing apparatus E in the fourth embodiment.

[0066] Figure 20 is a block diagram of the processing section E3 constituting the map expression data processing apparatus E.

[0067] Figure 21 is a flowchart illustrating an example of the operation of the map expression data processing apparatus E.

[0068] Figure 22 is a flowchart illustrating an example of the correspondence information acquisition processing.

[0069] Figure 23 is a flowchart illustrating an example of the distance relationship information acquisition processing.

[0070] Figure 24 is a flowchart illustrating an example of the angle relationship information acquisition processing.

[0071] Figure 25 is a flowchart illustrating an example of the region determination processing.

[0072] Figure 26 is a flowchart illustrating an example of the orientation information acquisition process.

[0073] Figure 27 is a flowchart illustrating an example of the site addition process.

[0074] Figure 28 is a flowchart illustrating an example of the output map configuration process.

[0075] Figure 29 is a diagram showing a specific processing example of the map expression data processing apparatus E.

[0076] Figure 30 is a diagram showing an example of the map expression data.

[0077] Figure 31 is a diagram showing an example of information acquired by the map expression data processing apparatus E in the course of processing.

[0078] Figure 32 is a diagram showing an example of information acquired by the map expression data processing apparatus E in the course of processing.

[0079] Figure 33 is a diagram showing an image of information acquired by the map expression data processing apparatus E in the course of processing.

[0080] Figure 34 is a diagram showing an example of information acquired by the map expression data processing apparatus E in the course of processing.

[0081] Figure 35 is a diagram showing an example of information acquired by the map expression data processing apparatus E in the course of processing.

[0082] Figure 36 is a diagram showing an image of information acquired by the map expression data processing apparatus E in the course of processing.

[0083] Figure 37 is a diagram showing an image of information acquired by the map expression data processing apparatus E in the course of processing.

[0084] Figure 38 is a diagram showing an example of information acquired by the map expression data processing apparatus E in the course of processing.

[0085] Figure 39 is a diagram showing an example of information acquired by the map expression data processing apparatus E in the course of processing.

[0086] Figure 40 is a diagram illustrating processing of the map expression data processing apparatus E.

[0087] Figure 41 is an image illustrating the processing of the map representation data processing apparatus E.

[0088] Figure 42 is a diagram showing an output example of the map representation data processing apparatus E.

[0089] Figure 43 is an image illustrating the processing of the map representation data processing apparatus E.

[0090] Figure 44 is an image illustrating the processing of the map representation data processing apparatus E.

[0091] Figure 45 is an image illustrating the processing of the map representation data processing apparatus E.

[0092] Figure 46 is a diagram showing an output example of the map representation data processing apparatus E.

[0093] Figure 47 is a diagram showing an overview of the computer system in the above embodiment.

[0094] Figure 48 is a block diagram of the computer system. DETAILED DESCRIPTION

[0095] Embodiments of a map representation data processing apparatus and the like will be described below with reference to the accompanying drawings. Furthermore, in the embodiments, constituent elements to which the same reference signs are assigned perform the same actions, and thus there are cases in which repeated description is omitted.

[0096] (First Embodiment)

[0097] In the present embodiment, a learning apparatus that learns two or more pieces of learning source information having a string having two or more characters and a label for classifying the string by a technique of machine learning, and constitutes a learner, is described. Furthermore, the label is a label related to the number of occurrences in a place dictionary, and is any of two or more labels. In addition, the label is preferably any of three kinds of 1, 0, and 2 or more in frequency of occurrence. Furthermore, the technique of machine learning is the same meaning as an algorithm using machine learning.

[0098] In addition, in the present embodiment, a learning apparatus that learns a learning module to which a string is assigned (a stream to which a string is assigned) in units of characters, and constitutes a learner, is described.

[0099] Further, in the present embodiment, a classification device that gives a string to a classification module by a learner constituted by a learning device and acquires a label corresponding to the string will be described. Further, the classification module can also be referred to as a prediction module.

[0100] Figure 1 is a block diagram of a learning device A in the present embodiment. In addition, Figure 2 is a block diagram of a classification device B in the present embodiment.

[0101] The learning device A has a storage section Al, a reception section A2, a processing section A3, and an output section A4. The storage section Al has a learning source information storage section Al l and a learner storage section A12. The processing section A3 has a learning section A31. The output section A4 has an accumulation section A41.

[0102] The classification device B has a storage section Bl, a reception section B2, a processing section B3, and an output section B4. The storage section Bl has the learner storage section A12. The reception section B2 has a string reception section B21. The processing section B3 has a classification section B31. The output section B4 has a label output section B41.

[0103] The storage section Al that constitutes the learning device A stores various information. The various information, for example, refers to learning source information to be described later, a learner to be described later, and a learning module. The learning module is a program that constitutes the learning section A31 and is a program for acquiring the learner. The learning module, for example, is a function in a machine learning architecture of fastText, tinySVM, TensorFlow, and the like, and various random forest functions and the like.

[0104] Two or more learning source information is stored in the learning source information storage section Al l. The learning source information is information of a learning object. The learning source information has a string and a label. The string has two or more characters. The label is information related to the number of occurrences in a place dictionary to be described later. The label can adopt information of any of two or more labels.

[0105] The label is preferably any of a first label indicating that the number of occurrences in the place dictionary is one time, a second label indicating that the number of occurrences in the place dictionary is zero times, and a third label indicating that the number of occurrences in the place dictionary is two or more times. That is, the label can preferably adopt information of any of three.

[0106] A place dictionary refers to a dictionary related to a place name. The place dictionary has, for example, two or more pieces of correspondence information having a place name and position information that specifies a position of a place determined by the place name. The place name refers to, for example, a name of a place, a name of a scenery, a name of a specific place, a name of a prefecture, a name of a city, town or village, a name of a river or a mountain or a park or a scenic spot, or the like. The place dictionary has, for example, two or more pieces of correspondence information having a place name and information that specifies a description of a place determined by the place name.

[0107] The learner is stored in the learner storage section A12. The learner can also be referred to as a classifier. In addition, the learner can also be referred to as a classification model, a prediction model, or the like. The learner is information obtained by a learning module using a technique of machine learning. The technique of machine learning is not limited to deep learning, SVM, decision tree, random forest, or the like. The technique of machine learning is preferably, for example, a technique using a recurrent neural network (RNN). That is, the learner is preferably an RNN.

[0108] The reception section A2 receives various information and instructions. Here, the reception refers to reception of input by a user, reception from an external device, or the like. However, it is sufficient that the various information and instructions can be obtained by the reception. The input unit of the various information and instructions can be, for example, any of an input unit based on a touch screen, a keyboard, a mouse, and a menu screen, or the like.

[0109] The processing section A3 performs various processing. The various processing is, for example, processing performed by the learning section A31.

[0110] The learning section A31 performs learning processing based on a technique of machine learning using two or more pieces of learning source information of the learning source information storage section A11, and obtains a learner.

[0111] The learning section A31, for example, sequentially gives each character of a character string possessed by each of the two or more pieces of learning source information that constitutes the learning source information storage section A11 to a learning module of machine learning, and gives a label paired with the character string to the learning module, executes the learning module, and obtains a learner. Further, the processing of sequentially giving each character that constitutes a character string to a learning module of machine learning can also be referred to as processing of giving a flow of a character string to a learning module of machine learning. That is, the learning section A31 can also be referred to as learning a flow of a character string possessed by each of the two or more pieces of learning source information. In addition, the learning module of machine learning is stored, for example, in the storage section A1.

[0112] In addition, the technique for acquiring machine learning of the learner can use, for example, deep learning, SVR, random forest, decision tree, and the like. In addition, the learning module of machine learning is, for example, a function in a machine learning architecture of fastText, tinySVM, TensorFlow, and the like, and various random forest functions, and the like.

[0113] In addition, the learner acquired by the learning unit A31 is information for outputting a label in a case where a classification module to which machine learning is applied is given a string as input. The learner acquired by the learning unit A31 is, for example, information for outputting a label in a case where characters constituting a string are sequentially given one character by one character.

[0114] Further, it can be considered that the learning unit A31 includes a learning module, and it can be considered that the learning unit A31 does not include a learning module.

[0115] The output unit A4 outputs various information. The various information is, for example, a learner. Here, the output generally refers to accumulation to a recording medium, but it can be considered to include display to a display, projection using a projector, printing by a printer, sound output, transmission to an external device, delivery of a processing result to another processing device and another program, and the like.

[0116] The accumulation unit A41 accumulates the learner acquired by the learning unit A31. The accumulation unit A41 generally accumulates the learner acquired by the learning unit A31 in the learner storage unit A12. However, the accumulation unit A41 can transmit the learner acquired by the learning unit A31 to an external device not illustrated.

[0117] The storage unit B1 constituting the classification device B stores various information. The various information is, for example, a learner, a classification module of machine learning, map information.

[0118] The technique of machine learning in the classification module of machine learning is not limited to deep learning, SVM, decision tree, random forest, and the like. The technique of machine learning is, for example, preferably a technique using a regression type neural network. The classification module is a program for classification of a string for which reception is accepted, and is, for example, a function in a machine learning architecture of fastText, tinySVM, TensorFlow, and the like, and various random forest functions, and the like.

[0119] The map information is information related to a map. The map information is, for example, map expression data. The map expression data is information that a geographical region is expressed. The map expression data is preferably associated with a map identifier that identifies the map expression data. The map identifier is, for example, an ID, a file name of the map expression data, a name of the map expression data, or the like. The map expression data is, for example, an old map, a pictorial map, a sketch map, a hand-drawn map, or the like, but the kind thereof is not limited. The map expression data is generally image data, but can also be vector data or the like, and the data structure is not limited. The map expression data is associated with one or more attribute values. The one or more attribute values are attribute values of the map expression data. The attribute values are information that represents the nature and characteristics of the map expression data or the like. The one or more attribute values associated with the map expression data include region determination information. The region determination information is information that determines a region expressed by the map expression data. The region is generally rectangular, but can also be a shape other than rectangular, such as triangular, octagonal, circular, or the like. The region expressed by the map expression data can also be referred to as a region represented by the map expression data. The region determination information is, for example, a set of (latitude, longitude). In addition, the region determination information is, for example, a set of information that represents relative coordinates from a reference point. However, the data structure of the region determination information is not limited as long as it is information that determines a region.

[0120] The one or more attribute values associated with the map expression data are, for example, static attribute values that are static. In addition, the one or more attribute values associated with the map expression data are, for example, dynamic attribute values that dynamically change. The one or more attribute values can include one or more static attribute values and one or more dynamic attribute values. The static attribute values are, for example, a scale of a map that is the map expression data (appropriately, referred to as simply "scale"), an actual area on a map in the region shown by the map expression data (appropriately, referred to as simply "area"), content information that represents the content of the map. The content information is the degree of completion of the map expression data, the theme of the map expression data, metadata corresponding to the map expression data, keywords, or the like. The metadata and the keywords are, for example, the kind of the map expression data, the name of a sightseeing spot existing in the region of the map expression data, a place name, or the like. The kind of the map expression data is, for example, "theme park" indicating a map of a theme park, "sightseeing map" indicating a map for sightseeing, information indicating a map of a specific region (for example, a school or the like), or the like. In addition, the one or more static attribute values can also be, for example, a mark indicating a position shown by explicit position determination information, or a mark indicating a position that cannot be indicated by the explicit position determination information, or the like. The dynamic attribute values are, for example, the distance from a position shown by the position determination information to a place that represents the map expression data, user action information related to the user's action with respect to the map expression data.

[0121] Further, the point representing the map expression data refers to, for example, a center point of the map expression data, an arbitrary point of an end portion of the map expression data, or an arbitrary point of a boundary of a region constituting the map expression data.

[0122] The map expression data is preferably contained in a file. However, the map expression data can also be data in a database, and the data format and management method thereof are not limited. In the case where the map expression data is contained in a file, two or more map expression data can also be contained in the file. Alternatively, one map expression data can also be realized by two or more files. That is, one map expression data can also be divided into two or more files.

[0123] The learner storage section A12 constituting the classification device B stores the learner accumulated by the learning device A.

[0124] The reception section B2 receives various information and instructions. Here, the reception refers to, for example, reception of input by a user, reception from an external device, and the like. However, as long as various information and instructions can be acquired through the reception, it is arbitrary.

[0125] The character string reception section B21 receives a character string. Here, the reception refers to, for example, a concept including reception of information input from an input device such as a keyboard and a mouse, a touch panel, and the like, reception of information transmitted through a wired or wireless communication line, reception of information read from a recording medium such as an optical disk and a magnetic disk, a semiconductor memory, and the like.

[0126] The character string reception section B21, for example, performs character recognition processing on map information, and acquires one or more character strings.

[0127] The processing section B3 performs various processing. The various processing refers to, for example, processing performed by the classification section B31.

[0128] The classification section B31 acquires a label corresponding to a character string using the learner of the learner storage section A12 and the character string received by the reception section B2 through a technique of machine learning.

[0129] The classification section B31, for example, sequentially gives two or more characters constituting the character string received by the reception section B2 to the classification module of machine learning in units of characters, and gives the learner of the learner storage section A12 to the classification module, executes the classification module, and acquires a label.

[0130] Further, it can be considered that the classification section B31 contains the classification module, and it can also be considered that the classification section B31 does not contain the classification module.

[0131] The output section B4 outputs various information. The various information is, for example, the label acquired by the classification section B31.

[0132] The label output section B41 outputs the label acquired by the classification section B31. Here, the output can be considered to include accumulation to a recording medium, display to a display, projection using a projector, printing by a printer, sound output, transmission to an external device, delivery of a processing result to other processing devices and other programs, and the like.

[0133] The storage section Al, the learning source information storage section Al l, the learner storage section Al 2, the storage section Bl, and the learner storage section Al 2 are preferably nonvolatile recording media, but can be realized by volatile recording media.

[0134] The process of storing information in the storage section Al and the like is not limited. For example, information can be stored in the storage section Al and the like by a recording medium, information transmitted through a communication line and the like can be stored in the storage section Al and the like, or information input through an input device can be stored in the storage section Al and the like.

[0135] The reception section A2, the reception section B2, and the character string reception section B21 can be realized by a device driver of an input unit such as a touch panel and a keyboard, and control software of a menu screen and the like.

[0136] The processing section A3, the learning section A31, the accumulation section A41, the processing section B3, and the classification section B31 can be generally realized by an MPU and a memory and the like. The processing steps of the processing section A3 and the like can be generally realized by software recorded in a recording medium such as a ROM. However, the processing steps can be realized by hardware (a dedicated circuit).

[0137] It can be considered that the output section A4, the output section B4, and the label output section B41 include output devices such as a display and a speaker, and it can be considered that the output section A4 and the like do not include the output devices. The output section A4 and the like can be realized by a driver software of the output devices, or the driver software of the output devices and the output devices and the like.

[0138] Next, an example of the operation of the learning device A will be described using the flowchart of Figure 3

[0139] (Step S301) The learning section A31 substitutes 1 for the counter i.

[0140] (Step S302) The learning section A31 determines whether the i-th learning source information exists in the learning source information storage section Al l. In the case where the i-th learning source information exists, the process proceeds to Step S303, and in the case where the i-th learning source information does not exist, the process proceeds to Step S312.

[0141] ​(Step S303) The learning unit A31 acquires the i-th learning source information from the learning source information storage unit Al l.

[0142] (Step S304) The learning unit A31 substitutes 1 for the counter j.

[0143] (Step S305) The learning unit A31 determines whether or not the j-th character exists in the string possessed by the i-th learning source information acquired in step S303. In the case where the j-th character exists, the process proceeds to step S306, and in the case where the j-th character does not exist, the process proceeds to step S309.

[0144] (Step S306) The learning unit A31 acquires the j-th character in the string possessed by the i-th learning source information acquired in step S303.

[0145] (Step S307) The learning unit A31 imparts the j-th character acquired in step S306 to the learning module of the storage unit Al. Further, the process of imparting the character to the learning module is broadly interpreted, for example, as imparting the character as a parameter of a function or a method, i.e., the learning module, imparting the character to the learning module as an execution module, and the like, and enables the learning module to perform a learning process using the character.

[0146] (Step S308) The learning unit A31 increments the counter j by 1. The process returns to step S305.

[0147] (Step S309) The learning unit A31 acquires the tag possessed by the i-th learning source information acquired in step S303.

[0148] (Step S310) The learning unit A31 imparts the tag acquired in step S309 to the learning module of the storage unit Al. Further, the process of imparting the tag to the learning module is broadly interpreted, for example, as imparting the tag as a parameter of a function or a method, i.e., the learning module, imparting the tag to the learning module as an execution module, and the like, and enables the learning module to perform a learning process using the tag.

[0149] (Step S311) The learning unit A31 increments the counter i by 1. The process returns to step S302.

[0150] (Step S312) The learning unit A31 executes the learning module to which the character is imparted, the tag, and acquires the learner.

[0151] (Step S313) The accumulation unit A41 accumulates the learner acquired in step S312 in the learner storage unit A12. The process ends.

[0152] Next, an example of the operation of the classification device B will be described using the flowchart of Figure 4

[0153] ​(Step S401) The character string receiving section B21 determines whether a character string has been received. If a character string has been received, the process proceeds to step S402, and if a character string has not been received, the process returns to step S401.

[0154] (Step S402) The classification section B31 assigns 1 to the counter i.

[0155] (Step S403) The classification section B31 determines whether the i-th character exists in the character string received in step S401. If the i-th character exists, the process proceeds to step S404, and if the i-th character does not exist, the process proceeds to step S407.

[0156] (Step S404) The classification section B31 acquires the i-th character in the character string received in step S401.

[0157] (Step S405) The classification section B31 imparts the i-th character acquired in step S404 to the classification module of the storage section B1. Furthermore, the process of imparting the character to the classification module is broadly interpreted, for example, as imparting the character as a parameter of the classification module as a function or a method, imparting the character to the classification module as an execution module, and the like, and enables the classification module to perform a classification process using the character.

[0158] (Step S406) The classification section B31 increments the counter i by 1. The process returns to step S403.

[0159] (Step S407) The classification section B31 imparts the learner of the learner storage section A12 to the classification module of the storage section B1. Furthermore, the process of imparting the learner to the classification module is broadly interpreted, for example, as imparting the learner as a parameter of the classification module as a function or a method, imparting the learner to the classification module as an execution module, imparting link information to the learner as a parameter of the classification module as a function or a method, imparting link information to the learner to the classification module as an execution module, and the like, and enables the classification module to perform a classification process using the learner.

[0160] (Step S408) The classification section B31 executes the classification module, and acquires a label.

[0161] (Step S409) The label output section B41 outputs the label acquired in step S408. The process returns to step S401.

[0162] Furthermore, in the flowchart of Figure 4 , the process is ended by power-off and interruption of the end of the process.

[0163] Hereinafter, the use of Figure 5The specific operation example of the learning device A and the classification device B in the present embodiment will be described. The processing of the learning device A is a learning stage 501. The processing of the classification device B is a classification stage 502.

[0164] Now, it is assumed that a plurality of learning source information having a string and a label is stored in the learning source information storage section A11 of the learning device A. Figure 6 The learning source information has a string and a label. The label is any of a first label (value "1") indicating that the number of occurrences in the place dictionary is one, a second label (value "0") indicating that the number of occurrences in the place dictionary is zero, and a third label (value "2") indicating that the number of occurrences in the place dictionary is two or more. Figure 6

[0165] Further, the learning device A gives the learning module of the storage section Al a plurality of learning source information in the learning stage 501, and configures a learner with the learning source information described in the above. Figure 6 Here, in the learning stage 501, the learning section A31 gives the learning module the character configuring the string possessed by the learning source information one character by one character in order. In addition, the learning section A31 gives the learning module the label possessed by the learning source information. Further, the learning section A31 executes the learning module, and acquires the learner. Further, the accumulation section A41 accumulates the learner 504 in the learner storage section A12. Here, the learning module is, for example, 503, and is, for example, a module using Long short-term memory (LSTM) in a regression type neural network.

[0166] Next, in the classification stage 502, the string reception section B21 of the classification device B performs a character recognition process on the map information of the storage section B1, for example, and acquires one or more strings. In addition, the process of performing a character recognition process on the map information as an image and acquiring one or more strings is a known technique, and thus detailed description is omitted.

[0167] Further, the classification section B31 gives the classification module the character configuring the string one or more strings in order, and gives the classification module the learner 504, executes the classification module, and acquires the score corresponding to the first label, the second label, and the third label for each string unit (505). Next, the classification section B31 acquires the label of the highest score. Next, the label output section B41 outputs the label of the highest score in association with the string (506).

[0168] According to the present embodiment, the learner for appropriately classifying the information related to the place name can be acquired.

[0169] ​In addition, according to the present embodiment, information related to a place name can be appropriately classified.

[0170] Further, the processing in the present embodiment can also be realized by software. Also, the software can be distributed by software download or the like. In addition, the software can be recorded in a recording medium such as a CD-ROM and distributed. Further, the same applies to other embodiments in the present specification. Further, the software that realizes the learning device A in the present embodiment is a program as described below. That is, the program is a program for causing a computer that can access a learning source information storage unit that stores two or more pieces of learning source information having a character string having two or more characters and a label that is a label related to the number of occurrences in a place dictionary and is any of two or more labels to function as the following: a learning unit that sequentially gives each character that constitutes the character string possessed by each piece of the two or more pieces of learning source information stored in the learning source information storage unit to a learning module that is machine-learned in character units and gives the label paired with the character string to the learning module, executes the learning module, and acquires a learner; and an accumulation unit that accumulates the learner.

[0171] In addition, the software that realizes the classification device B in the present embodiment is a program as described below. That is, the program is a program for causing a computer that can access a learner storage unit that stores a learner accumulated by the learning device A to function as the following: a character string reception unit that receives a character string; a classification unit that sequentially gives each character that constitutes the character string to a classification module that is machine-learned in character units and gives the learner to the classification module, executes the classification module, and acquires a label; and a label output unit that outputs the label.

[0172] (Second Embodiment)

[0173] In the present embodiment, a map expression data processing device that acquires one or more character strings from map expression data, acquires position information (latitude, longitude) paired with each of the one or more character strings, acquires coordinate information on the map expression data of each of the one or more character strings, and accumulatively stores the coordinate information of each of the one or more character strings in association with the position information is described.

[0174] Further, in the present embodiment, a map expression data processing apparatus that acquires one or more string from map expression data, acquires position information that corresponds to each of the one or more string, and stores coordinate information that corresponds to each of the one or more string in association with the position information is explained. Further, in the determination of the appropriate string, it is preferable to use the learner acquired by the learning apparatus A explained in the first embodiment. Further, in the determination of the appropriate string, it is preferable to use the classification processing in the classification apparatus B explained in the first embodiment.

[0175] Further, in the present embodiment, a map information generating apparatus that acquires one or more graphic feature points such as an intersection from map expression data, acquires position information that corresponds to each of the one or more graphic feature points, and stores coordinate information that corresponds to each of the one or more graphic feature points in association with the position information is explained. Further, the graphic feature points are preferably one or more of an intersection and a bridge.

[0176] Further, in the present embodiment, a map expression data processing apparatus that acquires two or more string from map expression data, acquires position information that corresponds to each of the two or more string, and stores coordinate information that corresponds to each of the string corresponding to the acquired position information in association with the position information is explained.

[0177] Further, in the present embodiment, a map expression data processing apparatus that determines a region in accurate map information that corresponds to a region of map expression data is explained. Further, in such a case, it is preferable to determine a region in accurate map information considering a region outside a string that does not exist in the map expression data.

[0178] Further, in the present embodiment, a map expression data processing apparatus that acquires orientation information of map expression data is explained.

[0179] Figure 7 is a block diagram of the map expression data processing apparatus C in the present embodiment. The map expression data processing apparatus C has a storage section C1, a reception section C2, a processing section C3, and an output section C4.

[0180] The storage section C1 includes a place dictionary storage section C11, a map expression data storage section C12, an accurate map information storage section C13, a learner storage section A12, and a correspondence information storage section C14. The reception section C2 includes a map expression data reception section C21. The processing section C3 includes a character string acquisition section C31, a classification section C32, a grouping section C33, a feature place detection section C34, a coordinate information acquisition section C35, a position information acquisition section C36, a region determination information acquisition section C37, and an orientation information acquisition section C38. The region determination information acquisition section C37 includes, for example, a first region determination information acquisition unit C371, an outside character string decision unit C372, a size information acquisition unit C373, a distance information acquisition unit C374, and a second region determination information acquisition unit C375. The output section C4 includes a correspondence output section C41, an accurate map information output section C42, and an orientation information output section C43.

[0181] Various information is stored in the storage section C1. The various information is, for example, a place dictionary to be described later, map expression data to be described later, accurate map information to be described later, a learner, correspondence information to be described later, and a classification module. Further, the storage section C1 can not include the place dictionary storage section C11. The place dictionary storage section C11 can be present in an external device not shown.

[0182] The place dictionary is stored in the place dictionary storage section C11. The place dictionary has one or more pieces of place position information. The place position information is information in which place information that specifies a place and position information that specifies a position are associated with each other. The place position information is, for example, information that has place information and position information. The place position information is, for example, information that has a pointer to the place information and a pointer to the position information. However, the data structure of the place position information is not limited.

[0183] Further, the place information is, for example, a place name, an ID that specifies a place. The place name is, for example, a name of a place, a name of a scenic spot, a name of a specified place, a name of a prefecture, a name of a city, town, or village, a name of a river or a mountain or a park or a scenic spot, a name of an intersection, or the like. Further, the position information is, for example, (latitude, longitude), but can be another data form. Further, the ID that specifies a place is usually associated with the place name.

[0184] One or more pieces of map expression data are stored in the map expression data storage section C12. The map expression data is data expressing a map. The map expression data is information limited to a certain geographical area. The map expression data is not accurate map information to be described later. The map expression data is, for example, data lacking accuracy as a map. The map expression data is, for example, an old map, a pictorial map, a sketch map, a hand-drawn map, or the like, but the kind thereof is not limited. The map expression data is preferably associated with a map identifier identifying the map expression data. The map identifier is, for example, an ID, a file name having the map expression data, a name of the map expression data, or the like.

[0185] The map expression data is generally image data, but can be vector data or the like, and the data structure is not limited. The map expression data can also include a string of place information. In addition, in the case where the map expression data is image data and vector data, it is generally data in which place information is expressed. In addition, the map expression data is preferably included in a file. The map expression data can also be a file. However, the map expression data can also be data in a database, and the data format and management method thereof are not limited. In the case where the map expression data is included in a file, two or more pieces of map expression data can also be included in the file. In addition, one piece of map expression data can also be realized by two or more files. That is, one piece of map expression data can also be divided into two or more files. In addition, of course, the map expression data can also be one file.

[0186] Accurate map information is stored in the accurate map information storage section C13. The accurate map information is information of an accurate electronic map. The kind of map expressed by the accurate map information, and the data form of the map information, and the like are not limited. The kind of map is, for example, a topographic map, and a relief map, a geological map, a land use map, a residential map, a route map, a road map, a guide map, an aerial photograph, a satellite photograph, or the like. In addition, the data form of the accurate map information is, for example, raster data, and vector data, a KIWI format, or the like.

[0187] One or more pieces of learner are stored in the learner storage section A12. The learner of the learner storage section A12 is preferably a learner acquired by the learning device A. In this case, as described above, the learner is information acquired by giving two or more pieces of learning source information having a string having two or more characters and a tag related to the number of occurrences in the place dictionary, which is an arbitrary one of two or more kinds of tags, to a learning module of machine learning, executing the learning module. However, the learner can also be a learner obtained by learning one or more positive examples of place name information and one or more negative examples of place name information by a technique of machine learning, not a learner acquired by the learning device A.

[0188] The correspondence information storage section C14 stores one or more pieces of correspondence information. The correspondence information is information indicating the correspondence of the coordinate information and the position information. The correspondence information can also be information of a group of the coordinate information and the position information. In addition, the correspondence information can also be information of a group of a pointer to the coordinate information and a pointer to the position information. The correspondence information can be associated with the place name information, or can have the place name information. The data structure of the correspondence information is not limited. The correspondence information is preferably associated with a map expression data identifier identifying the map expression data.

[0189] The reception section C2 receives various information and instructions. The various information and instructions are, for example, the map expression data, the action start instruction, the output instruction. The action start instruction is an instruction to start the action of acquiring the correspondence information of the map expression data processing apparatus C. The output instruction is an instruction to output the map expression data and the like. The output instruction has, for example, a map expression data identifier identifying the map expression data.

[0190] Here, the reception refers to the reception of the input of the user, the reception from an external apparatus (for example, a terminal apparatus not shown), and the like. However, as long as the various information and instructions can be acquired by the reception, it is acceptable. The input unit of the various information and instructions can be, for example, any of an input unit based on a touch panel, a keyboard, a mouse, and a menu screen, and the like.

[0191] The map expression data reception section C21 receives the map expression data. Here, the reception refers to the concept of the reception of the information input from the input device such as a keyboard and a mouse, a touch panel, and the like, the reception of the information transmitted through a wired or wireless communication line, the reception of the information read from a recording medium such as an optical disk and a magnetic disk, a semiconductor memory, and the like. That is, the map expression data reception section C21 can also read the map expression data from the map expression data storage section C12.

[0192] The processing section C3 performs various processing. The various processing is, for example, processing performed by the string acquisition section C31, the classification section C32, the grouping section C33, the feature place detection section C34, the coordinate information acquisition section C35, the position information acquisition section C36, the region determination information acquisition section C37, the orientation information acquisition section C38, the first region determination information acquisition unit C371, the outside string determination unit C372, the size information acquisition unit C373, the distance information acquisition unit C374, and the second region determination information acquisition unit C375.

[0193] The character string acquisition section C31 acquires one or more than two character strings from the map representation data received by the map representation data reception section C21. The character string acquisition section C31 preferably acquires three or more character strings from the map representation data. The character string acquisition section C31 acquires one or more character strings from the map representation data, for example, by character recognition processing. Further, the character recognition processing is a well-known technology, and thus detailed description is omitted. In addition, the character string acquisition section C31 acquires one or more character strings included in the map representation data, for example.

[0194] The classification section C32 determines whether each of one or more than two character strings acquired by the character string acquisition section C31 is a character string of a certain place.

[0195] The classification section C32 determines whether each of one or more than two character strings acquired by the character string acquisition section C31 is a character string of a place to be registered, for example. The character string of the place to be registered is place information or unique place information, for example.

[0196] The classification section C32 performs classification processing of each of one or more than two character strings acquired by the character string acquisition section C31 into one of place information and non-place information, for example, using a learner by a machine learning technique.

[0197] The classification section C32 performs learning processing by a machine learning technique using the character string of each of one or more than two place information as a positive example and the character string of each of one or more than two non-place information as a negative example, and performs processing of determining each of one or more than two character strings acquired by the character string acquisition section C31 into one of place information and non-place information using a learned learner.

[0198] Further, the machine learning technique is not limited to deep learning, SVM, decision tree, random forest, and the like as described above. In addition, the learning module that performs learning processing and constitutes a learner is a function in a machine learning architecture of fastText, tinySVM, TensorFlow, and the like, and various random forest functions, for example. Further, the classification module that performs determination processing is a function in a machine learning architecture of fastText, tinySVM, TensorFlow, and the like, and various random forest functions, for example.

[0199] The classification section C32 preferably acquires the label corresponding to each of one or more than two strings acquired by the string acquisition section C31 by the same process as the classification section B31. In this case, the classification section C32 sequentially gives each of two or more characters constituting the string acquired by the string acquisition section C31 to the machine-learned classification module in the character unit, and gives the learner of the learner storage section A12 to the classification module, executes the classification module, and acquires the label.

[0200] Further, the classification section C32, for example, preferably judges only the string corresponding to the first label as the certain place string. In addition, the classification section C32, for example, can also judge the string corresponding to the first label and the third label as the certain place string.

[0201] The grouping section C33 determines two or more pieces of position information indicating a position close to satisfying a condition decided in advance among three or more pieces of position information acquired by the position information acquisition section C36. That is, the grouping section C33 performs a process of excluding position information of a place not located in the map expression data. Further, one or more pieces of position information of a place not located in the map expression data is information indicating a position far to satisfying a condition decided in advance from other position information acquired by the position information acquisition section C36.

[0202] The process of the grouping section C33 is, for example, any of the following two processes.

[0203] (1) Process using representative position information

[0204] The grouping section C33, for example, determines representative position information as position information representing three or more pieces of position information acquired by the position information acquisition section C36. The representative position information is usually position information indicating a central part among three or more pieces of position information acquired by the position information acquisition section C36.

[0205] Next, the grouping section C33, for example, calculates a distance of each of position information other than the representative position information from the representative position information, and acquires position information other than the representative position information as position information far to satisfying a condition decided in advance. One or more pieces of position information other than the representative position information and the representative position information are position information determined by the grouping section C33. Further, position information not acquired here is excluded position information.

[0206] Further, examples of the acquisition method of the representative position information are (1-1) or (1-2) or (1-3) or (1-4) below.

[0207] (1-1) The grouping section C33, for example, calculates distances of three or more combinations of two position information out of three or more position information acquired by the position information acquisition section C36, and acquires two representative position information of a group corresponding to the shortest distance out of the distances of the three or more combinations.

[0208] (1-2) The grouping section C33 acquires any one of the shortest group acquired in the above (1-1) as the representative position information.

[0209] (1-3) The grouping section C33 can acquire a gravity point of a region including three or more position information acquired by the position information acquisition section C36 as the representative position information.

[0210] (1-4) The grouping section C33 can acquire position information of a position closest to a center point in a region including three or more position information acquired by the position information acquisition section C36 as the representative position information.

[0211] Further, the acquisition algorithm of the representative position information is not limited to the above (1-1) to (1-4).

[0212] (2) Clustering processing

[0213] The grouping section C33, for example, performs clustering processing of position information using three or more position information acquired by the position information acquisition section C36. Further, the grouping section C33 acquires two or more position information acquired by the clustering processing. Further, the clustering processing is a known technique, and thus detailed description is omitted.

[0214] Further, corresponding information corresponding to the position information decided by the grouping section C33 is accumulated. In addition, the corresponding information corresponding to the position information, for example, refers to corresponding information including the position information, corresponding information including a pointer to the position information.

[0215] The feature place detection section C34 detects a feature place of a place represented by a feature pattern which is a pattern decided in advance as a feature according to the map representation data. The feature place detection section C34, for example, detects a feature place of a place represented by a feature pattern which is a pattern decided in advance as a feature according to the map representation data by an image recognition technique. Further, the feature pattern is preferably one or more patterns of an intersection or a bridge.

[0216] For example, the feature pattern of the intersection or the bridge or the like is stored in the storage section Cl, and the feature place detection section C34, for example, extracts a region in accordance with the size of the feature pattern from the map representation data while shifting the pixel of the start point in the map representation data, calculates the similarity of the extracted region to the feature pattern of the storage section Cl, and detects a region in which the similarity is greater than or equal to a threshold value as a feature place.

[0217] In addition, the feature place detection section C34 performs processing of object recognition of the feature pattern with respect to the map expression data, for example, and detects the feature place. Further, the processing of object recognition is a publicly known technique and thus detailed description is omitted.

[0218] The coordinate information acquisition section C35 acquires coordinate information (normally, (x, y)) as information that determines a coordinate position corresponding to the string acquired by the string acquisition section C31 and as information that determines a relative coordinate position in the map expression data. The coordinate position corresponding to the string refers to a coordinate of a point that represents a region of the string in the map expression data, and is a representative point of the string. The representative point is a center point of the region of the string in the map expression data, a point at the upper left of the region of the string in the map expression data, a point at the lower right of the region of the string in the map expression data, or the like. That is, the coordinate position corresponding to the string acquired by the string acquisition section C31 can be a coordinate position of a center point of a region in which the string is arranged, can be a coordinate position of a place of a corner at the upper left of the region in which the string is arranged, can be a coordinate position of a place of a corner at the lower right of the region in which the string is arranged, or the like.

[0219] The coordinate information acquisition section C35 can also acquire coordinate information that determines a coordinate position corresponding to a string in the string acquired by the string acquisition section C31 that corresponds to the position information decided by the grouping section C33. In this case, it can also be said that the coordinate information acquisition section C35 acquires coordinate information that determines a coordinate position corresponding to the string acquired by the string acquisition section C31.

[0220] The coordinate information acquisition section C35 acquires coordinate information corresponding to the feature pattern. The coordinate information acquisition section C35 acquires information of a coordinate of a representative point of a region of the feature pattern detected by the feature place detection section C34.

[0221] The coordinate information acquisition section C35 acquires, as the feature place, a coordinate position of a center point of the region in the region of the feature pattern detected by the feature place detection section C34, for example. In addition, the coordinate information acquisition section C35 acquires, as the feature place, a coordinate position of a point at the upper left of the region in the region of the feature pattern in the map expression data, for example. In addition, the coordinate information acquisition section C35 acquires, as the feature place, a coordinate position of a point at the lower right of the region in the region of the feature pattern in the map expression data, for example.

[0222] The position information acquisition section C36 acquires position information corresponding to the position information of the place as a string acquired by the string acquisition section C31 using the place dictionary. The position information acquisition section C36 acquires position information corresponding to the position information of the place as a string acquired by the string acquisition section C31 from the place dictionary. Further, acquiring position information using the place dictionary generally means acquiring position information from the place dictionary.

[0223] The position information acquisition section C36 acquires position information corresponding to the position information of the place as a string corresponding to the feature pattern, for example, from the place dictionary. The string corresponding to the feature pattern refers to a string disposed at a position closest to the feature pattern, for example. The string corresponding to the feature pattern refers to a string in which a string disposed at a position closest to the feature pattern (for example, "ABC") is combined with the name of the feature pattern (for example, "intersection"), for example. That is, the position information acquisition section C36 acquires a string (for example, "ABC") on the map expression data as a string disposed at a position closest to the feature pattern, reads the name of the feature pattern (for example, "intersection") from the storage section Cl, synthesizes the two strings to acquire a string (for example, "ABC intersection"), and acquires position information corresponding to the string (for example, "ABC intersection") from the place dictionary storage section Cl l. Further, the position information acquisition section C36 can acquire position information from the place dictionary storage section Cl l present in an external device not shown.

[0224] The position information acquisition section C36 acquires position information corresponding to the position information of the place as a string corresponding to the feature pattern, for example, from the place dictionary. The string corresponding to the feature pattern refers to a string disposed at a position closest to the feature pattern, for example. The string corresponding to the feature pattern refers to a string in which a string disposed at a position closest to the feature pattern (for example, "ABC") is combined with the name of the feature pattern (for example, "intersection"), for example. That is, the position information acquisition section C36 acquires a string (for example, "ABC") on the map expression data as a string disposed at a position closest to the feature pattern, reads the name of the feature pattern (for example, "intersection") from the storage section Cl, synthesizes the two strings to acquire a string (for example, "ABC intersection"), and acquires position information corresponding to the string (for example, "ABC intersection") from the place dictionary storage section Cl l. Further, the position information acquisition section C36 can acquire position information from the place dictionary storage section Cl l present in an external device not shown.

[0225] The region determination information acquisition section C37 acquires region determination information that determines a region in the accurate map information as a region containing positions indicated by two or more pieces of position information decided by the grouping section C33.

[0226] The region determination information acquisition section C37 can acquire region determination information that determines a region in the accurate map information as a region containing positions indicated by two or more pieces of position information acquired by the position information acquisition section C36.

[0227] The region determination information acquisition section C37 acquires region determination information, for example, through the processing of the first region determination information acquisition unit C371, the outside string decision unit C372, the size information acquisition unit C373, the distance information acquisition unit C374, and the second region determination information acquisition unit C375.

[0228] The first region determination information acquisition unit C371 acquires first region determination information that determines a first region that is a region containing all the positions indicated by the two or more pieces of position information decided by the grouping section C33, and that is a region in the accurate map information. The first region determination information is, for example, position information that determines two points of a rectangle. However, the region determined by the first region determination information can not be a rectangle, but can be a polygon, a circle, or the like.

[0229] The first region determination information acquisition unit C371 preferably acquires first region determination information that determines a first region that is a rectangular region containing all the positions indicated by the two or more pieces of position information decided by the grouping section C33.

[0230] The first region determination information acquisition unit C371 preferably acquires first region determination information that determines a first region that is a rectangular region containing all the positions indicated by the two or more pieces of position information decided by the grouping section C33, and that has the smallest area.

[0231] The outside string decision unit C372 decides a string corresponding to each of one or more pieces of position information located at the outermost side, from among the position information acquired by the position information acquisition section C36. The outside string decision unit C372 preferably decides strings corresponding to four pieces of position information located at the four outermost sides (top, bottom, left, and right on the map expression data). The position information acquired by the position information acquisition section C36 can not be all the position information acquired by the position information acquisition section C36. The position information acquired by the position information acquisition section C36 can be the position information decided by the grouping section C33.

[0232] The size information acquisition unit C373 acquires size information that determines a size up to a corner of the map expression data, outside one or more strings decided by the outside string decision unit C372. The size information is, for example, the number of pixels. The size information is, for example, a size on coordinates.

[0233] The distance information acquisition unit C374 acquires distance information that determines a distance corresponding to the size information acquired by the size information acquisition unit C373, using two or more groups of the coordinate information acquired by the coordinate information acquisition section C35 and the position information acquired by the position information acquisition section C36. The distance information is information that determines an actual distance, a distance in the real world (information in units of m, km, or the like).

[0234] The distance information acquisition unit C374 acquires, for example, unit information (e.g., m / pixel) of information of a distance (e.g., m) in a unit of coordinate information (e.g., one pixel) using two or more sets of coordinate information (e.g., (xl, yl), (x2, y2)) acquired by the coordinate information acquisition section C35 and position information (e.g., (XI, Yl), (X2, Y2)) acquired by the position information acquisition section C36. That is, for example, the distance information acquisition unit C374 calculates the number of pixels A between (xl, yl) and (x2, y2) and the distance B of (XI, Yl) and (X2, Y2), and acquires the unit information by the calculation formula "unit information = distance B / number of pixels A". Further, the distance information acquisition unit C374, for example, multiplies the size information (e.g., the number of pixels) acquired by the size information acquisition unit C373 and the unit information (e.g., the distance for one pixel) to acquire the distance information. The distance information acquisition unit C374, for example, preferably acquires the distance information of the four sides.

[0235] The second region determination information acquisition unit C375 acquires second region determination information that determines a second region that is an expansion of the first region determined by the first region determination information, using the distance information acquired by the distance information acquisition unit C374. The second region determination information acquisition unit C375, for example, acquires second region determination information that determines a second region that is an expansion of the first region determination information to each side in accordance with each distance information of the four sides. The second region determination information is, for example, position information that determines two points of a rectangle. However, the region determined by the first region determination information can not be a rectangle, but can be a polygon, a circle, or the like.

[0236] The orientation information acquisition section C38 acquires orientation information related to the orientation of the map expression data using the region determination information or the second region determination information. In the case where the region determination information or the second region determination information is a rectangle, the orientation information acquisition section C38 preferably acquires orientation information of the angle of the figure of the rectangle of the region determination information or the second region determination information with respect to the orientation (e.g., true north) as a reference. The orientation information is, for example, the angle with respect to true north (information between 0 degrees and 360 degrees, or information between -180 degrees and 180 degrees). Further, the orientation information is, for example, information (e.g., a vector) that determines the direction of true north in the map expression data, or information that indicates the true upward direction from the center of the map expression data. The information that indicates the true upward direction from the center of the map expression data is, for example, an angle, for example, "0 degrees" in the case where the direction from the true upward direction of the center is true north, or "-90 degrees" or "270 degrees" in the case where the direction from the true upward direction of the center is true east.

[0237] The azimuth information acquisition section C38, for example, acquires a pair of the position information of two or more places from the position information acquired by the position information acquisition section C36. The azimuth information acquisition section C38 calculates a first angle, which is an angle of the two places with respect to a direction (for example, north) as a reference, from the two pieces of position information of the pair. Also, the azimuth information acquisition section C38 acquires, for example, the coordinate information of the pair acquired by the coordinate information acquisition section C35. Next, the azimuth information acquisition section C38 calculates a second angle, which is an angle in the map expression data, from the two pieces of coordinate information. Next, the azimuth information acquisition section C38 acquires the azimuth information related to the azimuth in the map expression data using the first angle and the second angle. In a case where the first angle is "0 degrees" (the two places are in a relationship of due north and due south) and the second angle is "90 degrees" (the values of the y-axes of the coordinate information of the two points are the same), the azimuth information acquisition section C38 acquires the azimuth information "90 degrees".

[0238] For example, in a case where the positional relationship indicated by each of the position information of the pair of place 1 and place 2 indicates due north, the first angle between place 1 and place 2 is calculated as 0 degrees. Also, the azimuth information acquisition section C38 acquires, for example, the coordinate information of the pair acquired by the coordinate information acquisition section C35. Next, the azimuth information acquisition section C38 acquires the azimuth information indicating the due north direction in the map expression data from the two pieces of coordinate information (for example, a vector from place 1 to place 2).

[0239] Further, the azimuth information acquisition section C38, for example, preferably acquires two or more pairs from the position information of three or more places, and performs the same processing for each of the two or more pairs to acquire two or more pieces of azimuth information, and calculates a representative value (for example, an average value or a median value, or the like) of the two or more pieces of azimuth information. The output section C4 outputs various information. The various information, for example, refers to the correspondence information, the map expression data, the accurate map information, and the azimuth information.

[0240] The correspondence output section C41 outputs the coordinate information acquired by the coordinate information acquisition section C35 and the position information acquired by the position information acquisition section C36 in a corresponding and associated manner. The output of the coordinate information and the position information in a corresponding and associated manner can also refer to the output of the correspondence information having the coordinate information and the position information. Here, the output generally refers to the accumulation to a recording medium (for example, the correspondence information storage section C14), but can also be considered to include the concept of display to a display, projection using a projector, printing by a printer, sound output, transmission to an external device, delivery of a processing result to another processing device and another program, and the like. Further, the correspondence output section C41 preferably also outputs the place information corresponding the coordinate information and the position information in pairs with the coordinate information and the position information.

[0241] The correspondence output section C41 outputs coordinate information corresponding to a string judged by the classification section C32 to be a string of a certain place in association with position information.

[0242] The correspondence output section C41 outputs coordinate information corresponding to a string classified by the classification section C32 as position information in association with position information.

[0243] The correspondence output section C41 outputs each of two or more pieces of position information decided by the grouping section C33 in association with each of two or more pieces of coordinate information acquired by the coordinate information acquisition section C35.

[0244] The accurate map information output section C42 outputs accurate map information, for example, in a manner of knowing a region determined by the region determination information. Here, it can be considered that outputting includes displaying to a display, projecting using a projector, storing to a recording medium, printing by a printer, sound output, transmitting to an external device, delivering a processing result to another processing device and another program, and the like. Further, the manner of knowing a region determined by the region determination information refers to, for example, surrounding the region determined by the region determination information with a frame, changing a background color of the region determined by the region determination information to another, and the like, but is not limited to these.

[0245] The accurate map information output section C42 outputs accurate map information, for example, preferably in a manner of knowing a region determined by the second region determination information.

[0246] The direction information output section C43 outputs direction information acquired by the direction information acquisition section C38. The output manner of the direction information is not limited. Here, it can be considered that outputting includes displaying to a display, projecting using a projector, storing to a recording medium, printing by a printer, sound output, transmitting to an external device, delivering a processing result to another processing device and another program, and the like.

[0247] The storage section C1, the place dictionary storage section C11, the map expression data storage section C12, the accurate map information storage section C13, the learner storage section A12, and the correspondence information storage section C14 are preferably non-volatile recording media, but can be realized by volatile recording media.

[0248] The process of storing information in the storage section C1 and the like is not limited. For example, information can be stored in the storage section C1 and the like by a recording medium, information transmitted through a communication line and the like can be stored in the storage section C1 and the like, or information input through an input device can be stored in the storage section C1 and the like.

[0249] The reception unit C2 and the map expression data reception unit C21 can also be implemented by a device driver of an input unit such as a touch panel and a keyboard, and control software of a menu screen, and the like.

[0250] The processing unit C3, the character string acquisition unit C31, the classification unit C32, the grouping unit C33, the feature place detection unit C34, the coordinate information acquisition unit C35, the position information acquisition unit C36, the region determination information acquisition unit C37, the orientation information acquisition unit C38, the first region determination information acquisition unit C371, the outside character string determination unit C372, the size information acquisition unit C373, the distance information acquisition unit C374, and the second region determination information acquisition unit C375 can also be implemented by an MPU and a memory, and the like. The processing procedure of the processing unit C3 and the like can be implemented by software, which is recorded in a recording medium such as a ROM. However, the processing unit C3 and the like can also be implemented by hardware (a dedicated circuit).

[0251] It can be considered that the output unit C4, the corresponding output unit C41, the accurate map information output unit C42, and the orientation information output unit C43 include an output device such as a display and a speaker, and it can also be considered that the output unit C4 does not include the output device. The output unit C4 can also be implemented by a driver software of the output device, or the driver software of the output device and the output device, and the like.

[0252] Next, an example of the operation of the map expression data processing apparatus C will be described using the flowchart of Figure 8

[0253] (Step S801) The reception unit C2 determines whether or not an operation start instruction has been received. In the case where the operation start instruction has been received, the process proceeds to Step S802, and in the case where the operation start instruction has not been received, the process proceeds to Step S821.

[0254] (Step S802) The processing unit C3 assigns 1 to the counter i.

[0255] (Step S803) The processing unit C3 determines whether or not the i-th map expression data exists in the map expression data storage unit C12. In the case where the i-th map expression data exists, the process proceeds to Step S804, and in the case where the i-th map expression data does not exist, the process returns to Step S801.

[0256] (Step S804) The map expression data reception unit C21 acquires the i-th map expression data from the map expression data storage unit C12.

[0257] (Step S805) The character string acquisition unit C31 acquires one or more character strings expressed on the i-th map expression data acquired in Step S804. The character string acquisition unit C31 acquires one or more character strings expressed on the i-th map expression data, for example, by character recognition processing. ​

[0258] (Step S806) The classification section C32 substitutes 1 for the counter j.

[0259] (Step S807) The classification section C32 judges whether or not the jth string exists in the strings acquired in step S805. In the case where the jth string exists, the process proceeds to step S808, and in the case where the jth string does not exist, the process proceeds to step S813.

[0260] (Step S808) The classification section C32 performs classification processing on the jth string, and acquires a tag. Further, the classification processing is, for example, the processing of the classification device B described above, and is, for example, the processing from step S402 to step S409 of the classification device B. Figure 4

[0261] (Step S809) The coordinate information acquisition section C35 judges whether or not the jth string is place information to be registered. If it is place information to be registered, the process proceeds to step S810, and if it is not place information to be registered, the process proceeds to step S812.

[0262] Further, the place information to be registered is, for example, a tag indicating that the string acquired in step S808 is a string to be registered. In addition, the tag indicating that the string is a string to be registered is, for example, the first tag.

[0263] (Step S810) The coordinate information acquisition section C35 acquires coordinate information of a representative point of a region on the ith map expression data, which is a region in which the jth string is arranged. Further, the representative point can be a center of gravity of the region, a point at the upper left, a point at the lower right, or the like. In addition, it is assumed that information of a region in which a string is arranged is acquired when the string is acquired from the map expression data.

[0264] (Step S811) The position information acquisition section C36 acquires position information paired with the place information of the jth string from the place dictionary storage section C11. Further, the corresponding output section C41 temporarily stores the position information paired with the coordinate information acquired in step S810 in at least a buffer not shown.

[0265] (Step S812) The classification section C32 increments the counter j by 1. The process returns to step S807.

[0266] (Step S813) The grouping section C33 performs grouping processing on two or more pieces of position information stored in step S811. For an example of the grouping processing, the flowchart of FIG. 14 is used for explanation. Figure 9

[0267] (Step S814) The corresponding output section C41 substitutes 1 for the counter k.

[0268] ​​(Step S815) The correspondence output section C41 determines whether or not the kth position information exists in the position information acquired in step S813. In the case where the kth position information exists, the process proceeds to step S816, and in the case where the kth position information does not exist, the process proceeds to step S818.

[0269] (Step S816) The correspondence output section C41 acquires the coordinate information paired with the kth position information from a buffer not shown. Further, the correspondence output section C41 stores the kth position information in association with the acquired coordinate information in the correspondence information storage section C14 in association with the ith map expression data.

[0270] (Step S817) The correspondence output section C41 increments the counter k by 1. The process returns to step S807.

[0271] (Step S818) The region determination information acquisition section C37 performs a region determination process. The region determination process is a process of determining a region in the accurate map information that is a region corresponding to the region of the ith map expression data. An example of the region determination process is explained using the flowchart of Figure 10 .

[0272] (Step S819) The orientation information acquisition section C38 performs an orientation acquisition process. The orientation acquisition process is a process of acquiring orientation information that determines the orientation of the ith map expression data. An example of the orientation acquisition process is explained using the flowchart of Figure 11 .

[0273] (Step S820) The processing section C3 increments the counter i by 1. The process returns to step S803.

[0274] (Step S821) The reception section C2 determines whether or not an output instruction has been received. In the case where the output instruction has been received, the process proceeds to step S822, and in the case where the output instruction has not been received, the process returns to step S801.

[0275] (Step S822) The output section C4 performs an output process. The process returns to step S801. Further, the output process is a process of outputting an instruction corresponding to the output instruction, such as the map expression data. An example of the output process is explained using the flowchart of Figure 12 .

[0276] Further, in the flowchart of Figure 8 , the process is ended by a power-off and an interruption of the process end.

[0277] Next, an example of the grouping process of step S813 is explained using the flowchart of Figure 9 .

[0278] (Step S901) The grouping section C33 assigns 1 to the counter i.

[0279] (Step S902) The grouping unit C33 determines whether there is an i-th group of two location information pieces among the two or more location information pieces obtained in step S811. If there is an i-th group of two location information pieces, proceed to step S903; otherwise, proceed to step S905.

[0280] (Step S903) The grouping unit C33 calculates the distance between the two points determined by the i-th pair of location information.

[0281] (Step S904) The grouping unit C33 increments the counter i by 1. Return to step S807.

[0282] (Step S905) The grouping unit C33 obtains two positional information points corresponding to the minimum distance between the two points calculated in step S903. Furthermore, the grouping unit C33 obtains one of the two positional information points as representative positional information.

[0283] (Step S906) The grouping unit C33 inputs 1 into the counter j.

[0284] (Step S907) The grouping unit C33 determines whether there is a j-th location information other than the location information obtained in step S811 among the two or more location information. If the j-th location information exists, proceed to step S908; otherwise, return to the previous processing step.

[0285] (Step S908) The grouping unit C33 calculates the distance between each location information other than the representative location information and the representative location information among the two or more location information obtained in step S811.

[0286] (Step S909) The grouping unit C33 determines whether the distance calculated in step S907 is close enough to meet a predetermined condition (e.g., whether the distance is less than a threshold, or whether the distance is less than or equal to a threshold). If the distance is close enough to meet the predetermined condition, proceed to step S910; otherwise, proceed to step S911.

[0287] (Step S910) The grouping unit C33 temporarily stores the j-th position information in a buffer (not shown). Furthermore, the position information temporarily stored in the buffer is the position information obtained by the grouping unit C33.

[0288] (Step S911) The grouping unit C33 increments the counter j by 1. Return to step S807.

[0289] Next, use Figure 10 The flowchart illustrates an example of the region determination process in step S818.

[0290] (Step S1001) The first region determination information acquiring unit C371 acquires the position information of the smallest longitude among the position information temporarily accumulated in the buffer not shown by the grouping section C33 in step S910.

[0291] (Step S1002) The first region determination information acquiring unit C371 acquires the position information of the largest longitude among the position information temporarily accumulated in the buffer not shown by the grouping section C33 in step S910.

[0292] (Step S1003) The first region determination information acquiring unit C371 acquires the position information of the smallest latitude among the position information temporarily accumulated in the buffer not shown by the grouping section C33 in step S910.

[0293] (Step S1004) The first region determination information acquiring unit C371 acquires the position information of the largest latitude among the position information temporarily accumulated in the buffer not shown by the grouping section C33 in step S910.

[0294] (Step S1005) The first region determination information acquiring unit C371 includes the four position information acquired in steps S1001 to S1004 in the boundary, acquires the region determination information having the position information of the four points of the corners of the region of the determined rectangle which determines the region containing all the position information acquired by the grouping section C33. Further, the first region determination information acquiring unit C371 preferably includes the four position information acquired in steps S1001 to S1004 in the boundary, acquires the region determination information having the position information of the four points of the corners of the region of the smallest rectangle which determines the region containing all the position information acquired by the grouping section C33.

[0295] (Step S1006) The outermost string deciding unit C372 acquires the strings of the four directions (up, down, left and right) of the outermost among the region determined by the region determination information. Also, the size information acquiring unit C373 acquires the size information of the number of the outermost pixels of the four respective strings of the outermost and the map expression data.

[0296] (Step S1007) The distance information acquiring unit C374 acquires the distance in the pixel unit (the unit of one coordinate) using two or more sets of position information and coordinate information.

[0297] (Step S1008) The distance information acquisition unit C374 multiplies the distance in pixel units obtained in step S1007 with the four size information obtained in step S1006 to calculate the distance expanded in each of the four directions. Furthermore, the four expansion directions are perpendicular to each of the four sides of the rectangle determined by the region determination information.

[0298] (Step S1009) The second region determination information acquisition unit C375 acquires second region determination information, which is information that determines the region whose width in the four directions of the first region determination information is extended accordingly with respect to the distance acquired in step S1008. Furthermore, the second region determination information typically defines two positional information points for a rectangular region.

[0299] (Step S1010) The second region determination information acquisition unit C375 stores the second region determination information obtained in step S1009 in association with the i-th map representation data. Return to the previous processing level.

[0300] Next, use Figure 11 The flowchart illustrates an example of the orientation acquisition process in step S819.

[0301] (Step S1101) The orientation information acquisition unit C38 acquires the second area determination information acquired by the second area determination information acquisition unit C375.

[0302] (Step S1102) The orientation information acquisition unit C38 uses the four location information constituting the second area determination information to acquire orientation information.

[0303] (Step S1103) The orientation information acquisition unit C38 stores the orientation information acquired in step S1102 in association with the i-th map representation data. Return to the previous processing stage.

[0304] Next, use Figure 12 The flowchart illustrates an example of the output processing in step S822.

[0305] (Step S1201) Output unit C4 obtains the map representation data identifier of the accepted output instruction.

[0306] (Step S1202) The output unit C4 obtains the map representation data identified by the map representation data identifier obtained in step S1201 from the map representation data storage unit C12.

[0307] (Step S1203) The output unit C4 obtains one or more coordinate information that are paired with the map representation data identifier obtained in step S1201 from the corresponding information storage unit C14.

[0308] (Step S1204) The outputter C4 acquires the orientation information paired with the map representation data acquired in step S1202.

[0309] (Step S1205) The outputter C4 constructs the map representation data that indicates the place determined by each coordinate information on the map representation data acquired by step S1202, and constructs the map representation data that indicates the orientation information acquired by step S1204.

[0310] (Step S1206) The outputter C4 acquires the second region determination information paired with the map representation data acquired in step S1202.

[0311] (Step S1207) The outputter C4 acquires the accurate map information of the region determined by the second region determination information acquired in step S1206 from the accurate map information storage C13.

[0312] (Step S1208) The outputter C4 acquires one or more position information paired with the map representation data identifier acquired in step S1201 from the correspondence information storage C14.

[0313] (Step S1209) The outputter C4 constructs the accurate map information that indicates the place determined by each position information acquired in step S1208 on the accurate map information acquired by step S1207.

[0314] (Step S1210) The outputter C4 outputs the map representation data acquired in step S1205. In addition, the outputter C4 outputs the accurate map information acquired in step S1209. The process returns to the parent process.

[0315] Further, in the flowchart of Figure 12 , only one of the map representation data and the accurate map information can be outputted.

[0316] In addition, in the flowchart of Figure 12 , the map representation data outputted can be the map representation data that does not indicate the place determined by each coordinate information. In addition, in the flowchart of Figure 12 , the map representation data outputted can be the map representation data that does not indicate the orientation information.

[0317] In addition, in the flowchart of Figure 12 , the accurate map information outputted can be the map representation data that does not indicate the second region determination information. In addition, in the flowchart of Figure 12 , the accurate map information outputted can be the map representation data that does not indicate the place determined by each position information. Further, in the flowchart of Figure 12In the flowchart, the accurate map information output can also be map representation data that explicitly defines the area.

[0318] The specific operation of the map representation data processing device C in this embodiment will be described below.

[0319] Now, for example, suppose Figure 13 Map representation data of K Street in City A, 1301 Figure 14 The map representation data of P Street 1401 is stored in the map representation data storage unit C12. Furthermore, it is assumed that map representation data of 1301 is stored in pairs with the map representation data identifier "K Street", and map representation data of 1401 is stored in pairs with the map representation data identifier "P Street".

[0320] In addition, the location dictionary storage unit C11 stores... Figure 15 The location dictionary shown. It stores two or more records with both "location information" and "location details". "Location information" is a meaningless string here, but is usually the location name. "Location details" is (latitude, longitude).

[0321] Furthermore, the corresponding information storage unit C14 stores information having Figure 16 The structure shown corresponds to an information management table. This table stores one or more records with "ID", "Map Representation Data Identifier", "Coordinate Information", "Location Information", and "Location Information". "ID" identifies the record. "Map Representation Data Identifier" identifies the map representation data. "Coordinate Information" represents the coordinate values ​​indicating the relative position on the map representation data. "Location Information" here is (latitude, longitude). "Location Information" is a meaningless string here, but is usually the location name.

[0322] In this situation, suppose the user inputs an action start instruction to the map display data processing device C.

[0323] Therefore, the receiving unit C2 of the map representation data processing device C begins its receiving action. Furthermore, the map representation data processing device C performs the aforementioned processing on the map representation data for "K Street" and the map representation data for "P Street," storing one or more corresponding pieces of information in the corresponding information management table in association with each piece of map representation data. The stored records are... Figure 16 The record with ID=1, 2, 3, ..., 38, ...

[0324] Furthermore, let's assume that the user, for example, inputs an output instruction with the map representation data identifier "K Street" into the map representation data processing device C.

[0325] Thus, the acceptance section C2 of the map expression data processing apparatus C accepts the output instruction. Next, the output section C4 performs the above-described output processing. Further, the output section C4 outputs the map expression data of 1301 and the accurate map information of 1302. Furthermore, the output section C4 outputs the map expression data of 1301 and the accurate map information of 1302. Figure 13 Figure 13

[0326] According to the present embodiment, the coordinate information of the place name and the like on the map expression data and the position information can be automatically acquired in association with each other.

[0327] According to the present embodiment, the appropriate character string of the place name and the like on the map expression data can be extracted, and the coordinate information of the place name and the like and the position information can be automatically acquired in association with each other.

[0328] According to the present embodiment, the appropriate character string of the place name and the like on the map expression data can be extracted with high precision, and the coordinate information of the place name and the like and the position information can be automatically acquired in association with each other.

[0329] According to the present embodiment, the graphic feature point on the map expression data can be acquired, and the coordinate information of the graphic feature point and the position information can be automatically acquired in association with each other.

[0330] According to the present embodiment, the range of the accurate map information corresponding to the map expression data can be specified.

[0331] According to the present embodiment, the appropriate range of the accurate map information corresponding to the map expression data can be specified.

[0332] Further, according to the present embodiment, the orientation information related to the orientation of the map expression data can be obtained.

[0333] Further, in the present embodiment, the essential configurations are the place dictionary storage section C11, the map expression data acceptance section C21, the character string acquisition section C31, the coordinate information acquisition section C35, the position information acquisition section C36, and the correspondence output section C41. Other configuration elements configuring the map expression data processing apparatus C can not exist.

[0334] Further, in the present embodiment, the output examples other than the above are Figure 13 Figure 14

[0335] ​​​​Further, the software that realizes the map expression data processing apparatus C in the present embodiment is a program as described below. That is, the program is a program for causing a computer that can access a place dictionary storage section that stores a place dictionary having one or more pieces of place information that are information that associates a place determined by a place name and a position determined by a position name, to function as each of the following sections: a map expression data reception section that receives map expression data of a kind that is any one or more of a pictorial map, a hand-drawn map, and an outline map; a character string acquisition section that acquires a character string from the map expression data; a coordinate information acquisition section that acquires coordinate information that is information that determines a coordinate position corresponding to the character string acquired by the character string acquisition section and that is information that determines a relative coordinate position in the map expression data; a position information acquisition section that acquires position information corresponding to the place information of the character string acquired by the character string acquisition section using the place dictionary; and a correspondence output section that outputs the coordinate information acquired by the coordinate information acquisition section in association with the position information acquired by the position information acquisition section.

[0336] (Third Embodiment)

[0337] In the present embodiment, a map expression data processing apparatus D that can indicate a current position on map expression data is described using the correspondence information acquired in the second embodiment.

[0338] Figure 17 is a block diagram of the map expression data processing apparatus D in the present embodiment.

[0339] The map expression data processing apparatus D has a storage section D1, a reception section D2, a processing section D3, and an output section D4. The storage section D1 has a map expression data storage section C12 and a correspondence information storage section C14.

[0340] The processing section D3 has a current position information acquisition section D31, a coordinate information acquisition section D32, and a data construction section D33. The output section D4 has a map expression data output section D41.

[0341] Various information is stored in the storage section D1. The various information is, for example, map expression data, correspondence information.

[0342] One or two or more pieces of map expression data are stored in the map expression data storage section C12.

[0343] Two or more pieces of correspondence information are stored in the correspondence information storage section C14. The correspondence information is preferably associated with map expression data.

[0344] The reception section D2 receives various information and instructions. The various information and instructions are, for example, an operation start instruction.

[0345] Here, the reception refers to the reception of the input of the user, the reception from an external device (for example, a terminal device not shown), and the like. However, as long as various information and instructions can be acquired through the reception, it is acceptable. The input unit of the various information and instructions can be, for example, any of an input unit based on a touch panel, a keyboard, a mouse, and a menu screen, and the like.

[0346] The processing section D3 performs various processing. The various processing is, for example, processing performed by the current position information acquisition section D31, the coordinate information acquisition section D32, and the data construction section D33.

[0347] The current position information acquisition section D31 acquires current position information that specifies a current position. The current position information acquisition section D31 can be implemented by, for example, a GPS receiver. The current position information acquisition section D31 is a well-known technology, and thus a detailed description is omitted.

[0348] The coordinate information acquisition section D32 acquires coordinate information corresponding to the current position information using two or more pieces of correspondence information paired with the map expression data of the object. The coordinate information is a coordinate value that specifies a relative position on the map expression data in the display.

[0349] The coordinate information acquisition section D32 reads, for example, correspondence information 1 (coordinate information 1 (xl, yl), position information 1 (XI, Yl)) and correspondence information 2 (coordinate information 2 (x2, y2), position information 2 (X2, Y2)) paired with the map expression data of the object from the correspondence information storage section C14. Further, the coordinate information acquisition section D32 acquires the current position information (X3, Y3) acquired by the current position information acquisition section D31. Next, the coordinate information acquisition section D32 calculates (XI - X2), (Yl - Y2) from the position information 1 (XI, Yl) and the position information 2 (X2, Y2). In addition, the coordinate information acquisition section D32 calculates (xl - x2), (yl - y2) from the coordinate information 1 (xl, yl) and the coordinate information 2 (x2, y2). Next, the coordinate information acquisition section D32 calculates the number of pixels with respect to a unit distance (for example, 100 m) in the x-axis from (XI - X2) and (xl - x2). In addition, the coordinate information acquisition section D32 calculates the number of pixels with respect to a unit distance (for example, 100 m) in the y-axis from (Yl - Y2) and (yl - y2). Next, the coordinate information acquisition section D32, for example, calculates the number of pixels with respect to a unit distance in the x-axis from (XI - X3), adds the number of pixels to xl, and acquires an x-coordinate (x3) with respect to the current position information. In addition, the coordinate information acquisition section D32, for example, calculates the number of pixels with respect to a unit distance in the y-axis from (Yl - Y3), adds the number of pixels to yl, and acquires a y-coordinate (y3) with respect to the current position information.

[0350] The data constructing section D33 constructs the current position additional map representation data on the map representation data as a map representation data which explicitly shows the position shown by the coordinate information acquired by the coordinate information acquiring section D32. Further, the explicit method of the position shown by the coordinate information is not limited. The data constructing section D33, for example, arranges a predetermined pattern at the position shown by the coordinate information.

[0351] The output section D4 outputs various information. The various information, for example, refers to the current position additional map representation data.

[0352] The map representation data output section D41 outputs the current position additional map representation data acquired by the data constructing section D33.

[0353] The storage section Dl, the map representation data storage section C12, and the correspondence information storage section C14 are preferably nonvolatile recording media, but can also be realized by volatile recording media.

[0354] The process of storing information in the storage section Dl and the like is not limited. For example, the information can be stored in the storage section Dl and the like by a recording medium, the information to be transmitted can be stored in the storage section Dl and the like by a communication line and the like, or the information to be input can be stored in the storage section Dl and the like by an input device.

[0355] The reception section D2 can be realized by a device driver of an input unit such as a touch panel and a keyboard, and a control software of a menu screen and the like.

[0356] The processing section D3, the current position information acquiring section D31, the coordinate information acquiring section D32, and the data constructing section D33 can also be generally realized by an MPU and a memory and the like. The processing procedure of the processing section D3 can be generally realized by a software, and the software can be recorded in a recording medium such as a ROM. However, it can also be realized by a hardware (a dedicated circuit).

[0357] It can be considered that the output section D4 and the map representation data output section D41 include an output device such as a display and a speaker, and it can also be considered that they do not include the output device. The output section D4 can be realized by a driving software of the output device, or the driving software of the output device and the output device and the like.

[0358] Next, an example of the operation of the map representation data processing apparatus D will be described using the flowchart of Figure 18

[0359] (Step S1801) The reception section D2 acquires the map representation data of the display object from the map representation data storage section C12.

[0360] (Step S1802) The current position information acquiring section D31 acquires the current position information. ​

[0361] The coordinate information acquiring section D32 acquires two or more pieces of correspondence information that correspond to the map representation data identifier that identifies the map representation data acquired in step S1801, from the correspondence information storage section C14. Next, the coordinate information acquiring section D32 acquires coordinate information that corresponds to the current position information, using the two or more pieces of correspondence information.

[0362] The data constructing section D33 constructs, on the map representation data acquired in step S1801, the current position additional map representation data that is the map representation data of the position indicated by the coordinate information acquired in step S1803.

[0363] The map representation data output section D41 outputs the current position additional map representation data constructed in step S1804. The process ends.

[0364] According to the present embodiment, the current position can be indicated on the map representation data using two or more pieces of automatically acquired correspondence information.

[0365] Furthermore, the software that realizes the map representation data processing apparatus D in the present embodiment is a program that causes a computer that can access a map representation data storage section that stores map representation data, and a correspondence information storage section that stores two or more pieces of correspondence information that are a group of coordinate information output by the map representation data processing apparatus C and position information acquired by the position information acquiring section, to function as the following sections: a current position information acquiring section that acquires current position information that specifies a current position; a coordinate information acquiring section that acquires coordinate information that corresponds to the current position information, using the two or more pieces of correspondence information; a data constructing section that constructs, on the map representation data, current position additional map representation data that is the map representation data of the position indicated by the coordinate information; and a map representation data output section that outputs the current position additional map representation data.

[0366] (Fourth Embodiment)

[0367] In the present embodiment, a map representation data processing apparatus is described that acquires three or more place names from map representation data, acquires coordinate information that corresponds to each of the place names from the map representation data, acquires real position information that corresponds to the place names from a place dictionary, and outputs correspondence information that is a group of appropriate coordinate information and position information, using the coordinate information and the position information.

[0368] Further, in the present embodiment, a map expression data processing apparatus that determines an inappropriate place based on relationship information related to a relationship between corresponding information between two places, and outputs map expression data of two or more pieces of corresponding information that do not include inappropriate corresponding information corresponding to the inappropriate place is described. Further, in the present embodiment, in order to determine an inappropriate place, for example, distance relationship information related to a distance, angle relationship information related to an angle is used. Further, an inappropriate place refers to a place corresponding to corresponding information that is considered to be erroneous.

[0369] Further, in the present embodiment, a map expression data processing apparatus that acquires and outputs scale information of map expression data is described.

[0370] Further, in the present embodiment, a map expression data processing apparatus that acquires and outputs a range of map expression data is described.

[0371] Further, in the present embodiment, a map expression data processing apparatus that appends a place name or the like in a range of map expression data to map expression data is described.

[0372] Further, in the present embodiment, a map expression data processing apparatus that acquires and outputs information related to an orientation of map expression data is described.

[0373] Figure 19 is a block diagram of the map expression data processing apparatus E in the present embodiment. Further, Figure 20 is a block diagram of the processing section E3 that constitutes the map expression data processing apparatus E.

[0374] The map expression data processing apparatus E is provided with a storage section C1, a reception section C2, a processing section E3, and an output section E4

[0375] The storage section C1 includes a place dictionary storage section C11, a map expression data storage section C12, an accurate map information storage section C13, a learner storage section A12, and a correspondence information storage section C14. The reception section C2 includes a map expression data reception section C21. The processing section E3 includes a character string acquisition section C31, a classification section C32, a feature place detection section C34, a coordinate information acquisition section C35, a position information acquisition section C36, a relationship information acquisition section E31, a correspondence information acquisition section E32, a scale acquisition section E33, a region determination information acquisition section E34, an additional place acquisition section E35, an additional position acquisition section E36, an additional coordinate acquisition section E37, a place registration section E38, and a direction information acquisition section E39. The correspondence information acquisition section E32 includes a judgment unit E321 and a correspondence information acquisition unit E322. The output section E4 includes a correspondence output section E41, a scale information output section E42, a region determination information output section E43, and a direction information output section E44.

[0376] The processing section E3 performs various processing. The various processing is, for example, processing performed by the character string acquisition section C31, the classification section C32, the feature place detection section C34, the coordinate information acquisition section C35, the position information acquisition section C36, the relationship information acquisition section E31, the correspondence information acquisition section E32, the scale acquisition section E33, the region determination information acquisition section E34, the additional place acquisition section E35, the additional position acquisition section E36, the additional coordinate acquisition section E37, the place registration section E38, the direction information acquisition section E39.

[0377] The relationship information acquisition section E31 acquires relationship information. The relationship information is information related to a relationship among three or more correspondence information. The relationship information acquisition section E31 acquires the relationship information using the three or more correspondence information. The relationship information is, for example, distance relationship information to be described later, angle relationship information to be described later. Further, the correspondence information is a group of coordinate information and position information corresponding to a character string acquired by the character string acquisition section C31. The coordinate information corresponding to the character string acquired by the character string acquisition section C31 is information acquired by the coordinate information acquisition section C35. The position information corresponding to the character string acquired by the character string acquisition section C31 from the map expression data is information acquired by the position information acquisition section C36. Further, the character string is usually a name of a place. The place is, for example, a station, a sightseeing spot, a scenic spot, a building, a prefecture, a city, a town, a village, a park, and the like.

[0378] The relationship information acquisition section E31 preferably acquires relative relationship information between correspondence information corresponding to two respective character strings among the three or more character strings acquired by the character string acquisition section C31, in units of a group of two character strings.

[0379] The relationship information acquiring section E31 acquires distance relationship information using the correspondence information of the two respective strings, for example, in units of a group of two strings. The distance relationship information is information indicating the relationship of the relative distance, which is the difference in the coordinate information of the two strings, and the absolute distance, which is the difference in the position information of the two strings. Further, the group of two strings refers to a group of two strings out of three or more strings acquired by the string acquiring section C31.

[0380] The distance relationship information is, for example, MPP. MPP is meters per pixel (m / pixel). In the case where the correspondence information corresponding to the two respective strings is correspondence information 1 "coordinate information 1 (xl, yl) position information 1 (XI, Yl)" and correspondence information 2 "coordinate information 2 (x2, y2) position information 2 (X2, Y2)", MPP = " (distance (m) of position information 1 and position information 2) / (number of pixels between coordinate information 1 and coordinate information 2)". In addition, the distance relationship information is, for example, PPM. PPM is "PPM = (number of pixels between coordinate information 1 and coordinate information 2) / (distance (m) of position information 1 and position information 2)". Further, the method of calculating the distance of position information 1 (XI, Yl) and position information 2 (X2, Y2) and the method of calculating the number of pixels of coordinate information 1 (xl, yl) and coordinate information 2 (x2, y2) are known techniques, and thus the description thereof is omitted here.

[0381] The relationship information acquiring section E31 acquires angle relationship information using the correspondence information of the two respective strings, for example, in units of a group of two strings.

[0382] The angle relationship information is information indicating the relationship of the relative angle, which is obtained from the coordinate information of the two strings, and the absolute angle. The relative angle is an angle of a straight line connecting the two coordinate information with respect to a reference line (for example, a straight line of the lower edge of the rectangular map expression data or a straight line of the right edge of the rectangular map expression data). In addition, the absolute angle is, for example, an angle of a straight line connecting a straight line of a direction (for example, east, north) serving as a reference and the position information of the two strings.

[0383] The angle relationship information is, for example, information related to the difference between the relative angle and the absolute angle (for example, "relative angle - absolute angle", "absolute angle - relative angle", "|relative angle - absolute angle|"). The information related to the difference between the relative angle and the absolute angle can also be, for example, (cos(relative angle - absolute angle), sin(relative angle - absolute angle)), (cos(absolute angle - relative angle), sin(absolute angle - relative angle)), or the like. Further, the information related to the difference between the relative angle and the absolute angle can also be appropriately referred to as Rotation.

[0384] The correspondence information acquisition section E32 uses the relationship information, and acquires only two or more pieces of correspondence information corresponding to the relationship information satisfying a predetermined relationship, among three or more pieces of correspondence information.

[0385] The correspondence information acquisition section E32 determines whether each of the three or more pieces of distance relationship information satisfies a predetermined relationship, and excludes correspondence information corresponding to distance relationship information not satisfying the predetermined relationship from the correspondence information corresponding to each of the three or more pieces of string acquired by the string acquisition section C31, and acquires only two or more pieces of remaining correspondence information.

[0386] The correspondence information acquisition section E32 determines whether each of the three or more pieces of angle relationship information satisfies a predetermined relationship, and excludes correspondence information corresponding to angle relationship information not satisfying the predetermined relationship from the correspondence information corresponding to each of the three or more pieces of string acquired by the string acquisition section C31, and acquires only two or more pieces of remaining correspondence information.

[0387] The correspondence information acquisition section E32 acquires only two or more pieces of correspondence information corresponding to the distance relationship information and the angle relationship information, preferably in a case where both of the distance relationship information and the angle relationship information satisfy a predetermined relationship.

[0388] The determination unit E321 constituting the correspondence information acquisition section E32 groups the three or more pieces of relationship information acquired by the relationship information acquisition section E31, and determines relationship information not entering a group. Further, as an algorithm for clustering relationship information, for example, hierarchical agglomerative clustering (HAC) can be used. For HAC, refer to Internet URLs “http: / / pub.ist.ac.at / ~edels / Papers / 1984-J-05-HierarchicalClustering.pdf” “http: / / citeseerx.ist.psu.edu / viewdoc / download?doi=10.1.1.299.7703&rep=rep1&type=pdf”. Further, the algorithm for clustering of a set of information is not limited.

[0389] The correspondence information acquisition unit E322 excludes correspondence information corresponding to relationship information not entering a group from the correspondence information corresponding to each of the three or more pieces of string acquired by the string acquisition section C31, and acquires only two or more pieces of remaining correspondence information.

[0390] The scale acquisition section E33 acquires a representative value of the distance relationship information from three or more distance relationship information, and acquires scale information of the map expression data using the representative value of the distance relationship information. Further, the representative value is, for example, an average value, a median value. For example, in a case where the representative value of the distance relationship information (here, for example, MPP) is α, the scale acquisition section E33 acquires scale information "α", for example.

[0391] The area determination information acquisition section E34 acquires a relative perpendicular distance as a perpendicular distance of coordinate information corresponding to the string acquired by the string acquisition section C31 from each of the four edges of the map expression data, and acquires area determination information that determines a range in the real world of the map expression data using the relative perpendicular distance and the scale information (for example, MPP). Further, the range in the real world refers to a range in accurate map information.

[0392] The area determination information acquisition section E34, for example, calculates a relative perpendicular distance of coordinate information possessed by two or more respective corresponding information finally acquired by the corresponding information acquisition section E32 from each edge (upper edge, lower edge, left edge, right edge) of the map expression data. Further, a technique of calculating a perpendicular distance of a point from a straight line is a publicly known technique.

[0393] Further, the area determination information acquisition section E34, for example, acquires a representative value (for example, a median value, an average value) of two or more relative perpendicular distances per edge unit. Next, the area determination information acquisition section E34, for example, acquires position information of a place with respect to the representative value of the relative perpendicular distance per edge unit. Next, the area determination information acquisition section E34, for example, acquires a latitude or a longitude of each edge by "distance relationship information (MPP) x representative value of the relative perpendicular distance" per edge unit.

[0394] The additional place acquisition section E35 acquires place information of the range indicated by the area determination information from the place dictionary. The additional place acquisition section E35, for example, preferably acquires two or more place information of the range indicated by the area determination information from the place dictionary, and acquires place information excluding the string acquired by the string acquisition section C31 from the acquired place information. Further, the place dictionary used can be the place dictionary of the place dictionary storage section C11, or a place dictionary existing in an external device not illustrated.

[0395] In addition, the additional place acquisition section E35 preferably acquires place information of the range indicated by the area determination information, and one or more attribute values (for example, a station name, whether or not a sightseeing spot, an intersection name, and the like) of the place information, and acquires place information paired with one or more attribute values satisfying the predetermined condition from the place dictionary.

[0396] The additional position acquisition unit E36 acquires position information corresponding to the position information acquired by the additional position acquisition unit E35 from the place dictionary.

[0397] The additional coordinate acquisition unit E37 acquires coordinate information corresponding to the position information acquired by the additional position acquisition unit E36. The additional coordinate acquisition unit E37, for example, acquires a difference in position information possessed by one corresponding information from the position information acquired by the additional position acquisition unit E36, uses the difference in position information and distance relationship information (MPP) to acquire a difference in coordinate information possessed by one corresponding information, uses the coordinate information possessed by one corresponding information and the difference in coordinate information to acquire coordinate information corresponding to the position information acquired by the additional position acquisition unit E36.

[0398] The place registration unit E38 registers the place information (string) acquired by the additional place acquisition unit E35 at the position indicated by the coordinate information acquired by the additional coordinate acquisition unit E37. Further, the place information is added to the map expression data.

[0399] The direction information acquisition unit E39 acquires a representative value of the angle relationship information from three or more angle relationship information, and acquires direction information indicating the direction of the map expression data using the representative value of the angle relationship information.

[0400] In the case where the angle relationship information (here, for example, "relative angle-absolute angle") is "-30", the north of the real world is the straight up direction in the map expression data, and the right up is an angle of 30 degrees. The direction information acquisition unit E39, for example, acquires direction information "30 degrees". Further, it is assumed that the direction information here is an angle of rotation clockwise from the straight up direction in the map expression data. Further, the direction information can also be a pattern corresponding to the acquired angle.

[0401] The output unit E4 outputs various information. The various information, for example, refers to corresponding information, scale information, map expression data, accurate map information, and direction information. The output here generally refers to accumulation to a recording medium, but can also be considered to include display to a display, projection using a projector, printing by a printer, sound output, transmission to an external device, delivery of a processing result to other processing devices and other programs, and the like.

[0402] The corresponding output unit E41 accumulates two or more corresponding information acquired by the corresponding information acquisition unit E32. The corresponding output unit E41 can also display the two or more corresponding information, or transmit to other devices.

[0403] The scale information output unit E42 outputs the scale information acquired by the scale acquisition unit E33. The output here generally refers to display, but can also be accumulation to a recording medium, transmission to an external device, and the like.

[0404] The region determination information output unit E43 outputs the region determination information acquired by the region determination information acquisition unit E34. This output typically refers to display, but it could also be storage on a recording medium, transmission to an external device, etc.

[0405] The orientation information output unit E44 outputs the orientation information acquired by the orientation information acquisition unit E39. This output typically refers to display, but it could also be storage on a recording medium, transmission to an external device, etc.

[0406] The processing unit E3, string acquisition unit C31, classification unit C32, feature location detection unit C34, coordinate information acquisition unit C35, location information acquisition unit C36, relationship information acquisition unit E31, correspondence information acquisition unit E32, scale acquisition unit E33, area determination information acquisition unit E34, additional location acquisition unit E35, additional location acquisition unit E36, additional coordinate acquisition unit E37, location recording unit E38, and orientation information acquisition unit E39 can typically be implemented using an MPU and memory, etc. The processing steps of the processing unit E3, etc., are usually implemented in software, which is recorded on a recording medium such as ROM. However, they can also be implemented in hardware (dedicated circuitry).

[0407] The output unit E4, corresponding output unit E41, scale information output unit E42, area determination information output unit E43, and orientation information output unit E44 can be considered to include output devices such as displays and speakers, or they can be considered not to include them. The output unit E4, etc., can also be implemented through the driver software of the output device, or the driver software of the output device and the output device, etc.

[0408] Next, use Figure 21 The flowchart illustrates the operation of the map representation data processing device E. Figure 21 In the flowchart, for the... Figure 8 The same steps as in the flowchart are omitted. Furthermore, in Figure 21 The flowchart includes steps S804 to S812.

[0409] (Step S2101) Processing unit E3 uses the map representation data obtained in step S804 to perform corresponding information acquisition processing. For an example of corresponding information acquisition processing, it uses... Figure 22 The flowchart is explained below. Furthermore, the corresponding information acquisition process involves acquiring one or more accumulated corresponding information. Additionally, the map representation data being processed is typically the map representation data stored in the map representation data storage unit C12, but it can also be map representation data received from an external device.

[0410] (Step S2102) The correspondence output section E41 accumulates the one or more pieces of correspondence information acquired in step S2101. Further, the accumulation destination of the correspondence information is, for example, the storage section Cl, but can be another device. The correspondence output section E41 accumulates the one or more pieces of correspondence information acquired in step S2101 in association with the map expression data acquired in step S804.

[0411] (Step S2103) The scale acquisition section E33 acquires scale information using the representative value (for example, MPP) of the distance relationship information acquired in step S2101, and accumulates the scale information in association with the map expression data acquired in step S804. Further, the accumulation destination of the scale information is, for example, the storage section Cl, but can be another device.

[0412] (Step S2104) The region determination information acquisition section E34 performs a region determination process. For an example of the region determination process, a flowchart of Figure 25 is used for explanation. Further, the region determination process refers to a process of acquiring information indicating a region of the real world corresponding to a map region of the map expression data.

[0413] (Step S2105) The orientation information acquisition section E39 performs an orientation information acquisition process. For an example of the orientation information acquisition process, a flowchart of Figure 26 is used for explanation. Further, the orientation information acquisition process refers to a process of acquiring information related to the orientation (may be referred to as tilt) of the map expression data.

[0414] (Step S2106) The point addition section E35 and the like perform a point addition process. For an example of the point addition process, a flowchart of Figure 27 is used for explanation. The point addition process refers to a process of adding point information (string) not recorded in the map expression data.

[0415] (Step S2107) The processing section E3 performs a process of constructing an output map. For an example of such an output map construction process, a flowchart of Figure 28 is used for explanation.

[0416] (Step S2108) The output section E4 outputs the map expression data constructed in step S2107, and the accurate map information constructed in step S2107. The process ends. Further, here, the output section E4 can perform output of only one of the map expression data and the accurate map information.

[0417] Next, an example of the correspondence information acquisition process of step S2101 is explained using a flowchart of Figure 22 .

[0418] The relationship information acquisition section E31 performs distance relationship information acquisition processing. An example of the distance relationship information acquisition processing is explained using the flowchart of Fig. 23. Further, the distance relationship information acquisition processing is processing of acquiring one or more pieces of distance relationship information for a set of two or more pieces of correspondence information. Figure 23

[0419] The judgment section E321 groups three or more pieces of distance relationship information acquired in step S2201. Further, the judgment section E321 acquires one or two or more groups of distance relationship information. Further, for example, the HAC described above is used in the grouping.

[0420] The correspondence information acquisition section E322 assigns 1 to the counter i.

[0421] The correspondence information acquisition section E322 judges whether or not the i-th group acquired by step S2202 exists. In the case where the i-th group exists, the process proceeds to step S2205, and in the case where the i-th group does not exist, the process returns to the higher-level process.

[0422] The correspondence information acquisition section E322 judges whether or not the i-th group satisfies a predetermined condition related to the number of places within the group. In the case where the condition is satisfied, the process proceeds to step S2206, and in the case where the condition is not satisfied, the process proceeds to step S2209. Further, the predetermined condition is, for example, the number of places (may also be referred to as strings) corresponding to the i-th group is greater than or equal to a threshold value, the number of places corresponding to the i-th group is greater than the threshold value, the proportion of places corresponding to the i-th group is greater than or equal to a threshold value, the proportion of places corresponding to the i-th group is greater than a threshold value, or the like. Further, the proportion means "the number of places corresponding to the i-th group / the number of places belonging to any group".

[0423] The correspondence information acquisition section E322 acquires correspondence information for two or more places corresponding to the i-th group.

[0424] The relationship information acquisition section E31 performs angle relationship information acquisition processing for the correspondence information for two or more places acquired in step S2206. An example of the angle relationship information acquisition processing is explained using the flowchart of Fig. 24. Figure 24

[0425] ​​(Step S2208) The judging unit E321 groups the three or more angle relationship information acquired in step S2207. Further, the judging unit E321 acquires one or two or more groups of the angle relationship information. In addition, the grouping uses, for example, the HAC described above.

[0426] (Step S2209) The correspondence information acquiring unit E322 substitutes 1 to the counter j.

[0427] (Step S2210) The correspondence information acquiring unit E322 judges whether or not the jth group exists in the groups acquired in step S2208. In the case where the jth group exists, the process proceeds to step S2211, and in the case where the jth group does not exist, the process proceeds to step S2214.

[0428] (Step S2211) The correspondence information acquiring unit E322 judges whether or not the jth group satisfies a predetermined condition related to the number of places within the group. In the case where the condition is satisfied, the process proceeds to step S2212, and in the case where the condition is not satisfied, the process proceeds to step S2213. Further, the predetermined condition is, for example, that the number of places corresponding to the jth group (may be referred to as a string) is greater than or equal to a threshold value, that the number of places corresponding to the jth group is more than a threshold value, that the proportion of places corresponding to the jth group is greater than or equal to a threshold value, that the proportion of places corresponding to the jth group is greater than a threshold value, and the like. Further, the proportion means "the number of places corresponding to the jth group / the number of places belonging to any group".

[0429] (Step S2212) The correspondence information acquiring unit E322 acquires correspondence information of two or more places corresponding to the jth group.

[0430] (Step S2213) The correspondence information acquiring unit E322 increments the counter j by 1. The process returns to step S2210.

[0431] (Step S2214) The correspondence information acquiring unit E322 increments the counter i by 1. The process returns to step S2204.

[0432] Further, in the flowchart of Figure 22 , in step S2202 or step S2208, in the case where a plurality of groups are detected, the correspondence information acquiring unit E322 can select only one group having the largest number of places.

[0433] Next, using the flowchart of Figure 23 , an example of the distance relationship information acquiring process of step S2201 will be described.

[0434] (Step S2301) The relationship information acquiring unit E31 substitutes 1 to the counter i.

[0435] (Step S2302) The relationship information acquisition section E31 determines whether or not there is a group of the i-th two places among the three or more places to be processed. In the case where there is the group of the i-th two places, the process proceeds to step S2303, and in the case where there is not, the process returns to the process of the upper level.

[0436] (Step S2303) The relationship information acquisition section E31 acquires position information of the two respective places of the i-th group, and calculates an absolute distance of the two places from the two pieces of position information.

[0437] (Step S2304) The relationship information acquisition section E31 acquires coordinate information of the two respective places of the i-th group, and calculates a relative distance of the two places from the two pieces of coordinate information.

[0438] (Step S2305) The relationship information acquisition section E31 acquires distance relationship information (for example, MPP) using the absolute distance calculated in step S2303 and the relative distance calculated in step S2304, and temporarily stores the distance relationship information in association with the information of the i-th group.

[0439] (Step S2306) The relationship information acquisition section E31 increments the counter i by 1. The process returns to step S2302.

[0440] Next, an example of the angle relationship information acquisition process of step S2207 will be described using the flowchart of Figure 24

[0441] (Step S2401) The relationship information acquisition section E31 assigns 1 to the counter i.

[0442] (Step S2402) The relationship information acquisition section E31 determines whether or not there is a group of the i-th two places among the three or more places to be processed. In the case where there is the group of the i-th two places, the process proceeds to step S2403, and in the case where there is not, the process returns to the process of the upper level.

[0443] (Step S2403) The relationship information acquisition section E31 acquires position information of the two respective places of the i-th group, and calculates an absolute angle of the two places from the two pieces of position information.

[0444] (Step S2404) The relationship information acquisition section E31 acquires coordinate information of the two respective places of the i-th group, and calculates a relative angle of the two places from the two pieces of coordinate information.

[0445] ​(Step S2405) The relationship information acquisition unit E31 uses the absolute angle calculated in step S2403 and the relative angle calculated in step S2404 to acquire angle relationship information (e.g., Rotation), and temporarily stores the angle relationship information in association with the information of the i-th group.

[0446] (Step S2406) The relation information acquisition unit E31 increments the counter i by 1. Return to step S2402.

[0447] Next, use Figure 25 The flowchart illustrates an example of the region determination process in step S2104.

[0448] (Step S2501) The region determination information acquisition unit E34 inputs 1 into the counter i.

[0449] (Step S2502) The region determination information acquisition unit E34 determines whether there is an edge for the i-th processing object. If the i-th edge exists, proceed to step S2503; otherwise, return to the previous processing step. Furthermore, map data is typically represented as rectangles, and the edges of the processing object are usually the top, bottom, left, and right edges.

[0450] (Step S2503) The region determination information acquisition unit E34 acquires the edge information of the i-th edge of the determined map representation data. Furthermore, the edge information is, for example, the coordinate information (x1, y1) and (x2, y2) of the two endpoints constituting the edge. Additionally, the region determination information acquisition unit E34 may acquire the coordinate information of the two endpoints of the edge of the map representation data file, for example, assuming the entire map representation data (usually an image file) is considered as map data. Furthermore, the region determination information acquisition unit E34 may also perform contour extraction processing on the map representation data to acquire information about the contour lines of areas outside the map (e.g., blank areas) and the map area. The process of acquiring such contour line information is the process of acquiring edge information.

[0451] (Step S2504) The region determination information acquisition unit E34 inputs 1 into the counter j.

[0452] (Step S2505) The area determination information acquisition unit E34 determines whether there is information about the j-th object of the processing object. If the information about the j-th object exists, proceed to step S2506; otherwise, proceed to step S2510. Furthermore, the object information of the processing object corresponds to its location.

[0453] (Step S2506) The region determination information acquisition unit E34 acquires the information of the j-th object of the processing object.

[0454] (Step S2507) The region determination information acquisition unit E34 uses the coordinate information of the j-th object information and the edge information of the i-th edge to calculate the straight-line distance (e.g., number of pixels) between the j-th location and the i-th edge in the map representation data.

[0455] (Step S2508) The region determination information acquisition unit E34 uses the position information of the j-th object information, the straight-line distance calculated in step S2507, and the representative value of the distance relationship information (e.g., MPP) to obtain the position information of the two endpoints of the i-th edge. Furthermore, the position information of these two endpoints is used as edge information.

[0456] (Step S2509) The region determination information acquisition unit E34 increments the counter j by 1. Return to step S2505.

[0457] (Step S2510) The region determination information acquisition unit E34 acquires representative values ​​(e.g., median, average) for two or more edge information for the i-th edge. Furthermore, such representative values ​​are representative edge information. Representative edge information is, for example, information about the midline of two or more edge information. Representative edge information is, for example, the average of the two endpoints of each of the two or more edge information. For example, in two edge information 1 (x... 11 y 11 ), (x 12 y 12 ), edge information 2 (x 21 y 21 ), (x 22 y 22 In the case of ), the representative edge information of the average value is ((x) 11 +x 21 / 2), (y 11 +y 21 / 2)), ((x 12 +x 22 / 2), (y 12 +y 22 / 2)).

[0458] (Step S2511) The region determination information acquisition unit E34 increments the counter i by 1. Return to step S2502.

[0459] Next, use Figure 26 The flowchart illustrates an example of obtaining and processing the orientation information in step S2105.

[0460] (Step S2601) The orientation information acquisition unit E39 inputs 1 into the counter i.

[0461] (Step S2602) The azimuth information acquisition section E39 determines whether or not there is a group of the i-th place. In the case where there is a group of the i-th place, the process proceeds to step S2603, and in the case where there is not a group of the i-th place, the process proceeds to step S2605.

[0462] (Step S2603) The azimuth information acquisition section E39 acquires the angle relationship information of the i-th place.

[0463] (Step S2604) The azimuth information acquisition section E39 increments the counter i by 1. The process returns to step S2602.

[0464] (Step S2605) The azimuth information acquisition section E39 acquires a representative value (for example, an average value, a median value) from among the two or more angle relationship information acquired in step S2603.

[0465] (Step S2606) The azimuth information acquisition section E39 acquires the angle information using the representative value from among the angle relationship information acquired in step S2605. The process returns to the parent process. Further, the azimuth information acquisition section E39 acquires the angle information according to the calculation formula "angle information = f (representative value of angle relationship information)". Further, the calculation formula is, for example, "angle information = -1 x (representative value of angle relationship information)".

[0466] Next, an example of the place addition process of step S2106 will be described using the flowchart of Figure 27

[0467] (Step S2701) The place addition section E35 assigns 1 to the counter i.

[0468] (Step S2702) The place addition section E35 determines whether or not there is the i-th place position information in the place dictionary. In the case where there is the i-th place position information, the process proceeds to step S2703, and in the case where there is not the i-th place position information, the process returns to the parent process.

[0469] (Step S2703) The position addition section E36 acquires the position information possessed by the i-th place position information from the place dictionary.

[0470] (Step S2704) The place addition section E35 determines whether or not the position information acquired in step S2703 is within the range of the region indicated by the region determination information. In the case where it is within the range of the region, the process proceeds to step S2705, and in the case where it is not within the range of the region, the process proceeds to step S2711. Further, in the case where there is the region determination information indicating the range of the region, it is known technology to determine whether or not one position information is position information of a place within the region.

[0471] ​(Step S2705) The additional coordinate acquisition section E37 acquires coordinate information corresponding to the position information acquired in step S2703.

[0472] (Step S2706) The additional place acquisition section E35 acquires place information possessed by the i-th place position information from the place dictionary.

[0473] (Step S2707) The additional place acquisition section E35 judges whether or not to arrange the place information acquired in step S2706 in the map expression data. In the case of arrangement, it proceeds to step S2708, and in the case of non-arrangement, it proceeds to step S2709. Further, the additional place acquisition section E35 judges to arrange in the map expression data, for example, in the case where one or more attribute values corresponding to the place information acquired in step S2706 satisfy a predetermined condition.

[0474] (Step S2708) The place recording section E38 arranges the string of the place information acquired in step S2706 at the position shown by the coordinate information acquired in step S2705 on the map expression data.

[0475] (Step S2709) The correspondence information acquisition section E32 constructs correspondence information having the correspondence information acquired in step S2703 and the coordinate information acquired in step S2705.

[0476] (Step S2710) The correspondence output section E41 accumulates the correspondence information constructed in step S2709.

[0477] (Step S2711) The counter i is incremented by 1. It returns to step S2702.

[0478] Next, an example of the output map construction process of step S2107 will be described using the flowchart of Figure 28

[0479] (Step S2801) The output section E4 acquires the map expression data which is the object of the process. Further, the map expression data here is preferably data in which a place has been added according to the above-described place addition process.

[0480] (Step S2802) The processing section E3 acquires the current position information. Further, the processing section E3 can also have the current position information acquisition section D31.

[0481] (Step S2803) The processing section E3 acquires coordinate information corresponding to the current position information acquired in step S2802 using two or more correspondence information. Further, the processing section E3 can also have the coordinate information acquisition section D32.

[0482] ​(Step S2804) The processing section E3 constructs the current position indicated by the position indicated by the coordinate information acquired in step S2803 on the map expression data to which the map expression data is additionally attached. Further, the processing section E3 can also have a data construction section D33.

[0483] (Step S2805) The processing section E3 acquires the region determination information acquired by the above-described region determination processing.

[0484] (Step S2806) The processing section E3 acquires accurate map information including the region indicated by the region determination information acquired in step S2805.

[0485] (Step S2807) The processing section E3 constructs accurate map information indicating the range of the region indicated by the region determination information. The processing returns to the parent processing.

[0486] Further, in the flowchart of Figure 28 , the bearing information and the scale information are attached to the current position attached map expression data.

[0487] Hereinafter, a specific example of the operation of the map expression data processing apparatus E in the present embodiment will be described.

[0488] In the present specific example, the map expression data processing apparatus E performs the processing shown in Figure 29 . That is, the map expression data processing apparatus E sequentially performs (1) an operation based on an OCR module, (2) an operation based on a machine learning module, (3) an operation based on a geocoding module, (4) an operation based on a clustering module, and (5) an operation based on a map information extraction module. Further, in Figure 29 , the processing of (2) is optional.

[0489] Now, assume that the map expression data shown in Figure 30 is stored in the map expression data storage section C12. Further, the map expression data can also be appropriately referred to as an input image.

[0490] In such a situation, assume that the map expression data reception section C21 reads the input image from the map expression data storage section C12.

[0491] Assume that the character string acquisition section C31 performs character recognition processing (OCR) on the input image, and acquires a list of character strings "A post office", "B station", "2015", "C civic plaza", "D hospital branch", and "cargo terminal".

[0492] In addition, the coordinate information acquisition section C35 acquires coordinate information of each of the strings acquired by the string acquisition section C31. The coordinate information is relative coordinates (x, y) in the input image. The coordinate information acquisition section C35 acquires, for example, coordinate information of a barycentric point of a region in which the string is arranged.

[0493] Next, the classification section C32 determines whether each of the strings acquired by the string acquisition section C31 is place information using a technique of machine learning (here, LSTM Neural Network), and writes a place information flag "-1" for "2015" which is not place information. According to the above processing, for example, information of Figure 31 can be obtained. In addition, the place information flag refers to a flag for determining whether a string is place information.

[0494] Next, the position information acquisition section C36 acquires position information (latitude, longitude) corresponding to each of the strings "A post office", "B station", "C citizens square", "D hospital branch", and "cargo terminal" by searching a place dictionary using each of the strings as a keyword. Also, for example, information of Figure 32 can be obtained.

[0495] Next, the relationship information acquisition section E31 performs a check processing using distance relationship information as follows. That is, the relationship information acquisition section E31 calculates a relative distance (here, the number of pixels) of a difference in coordinate information between places belonging to each of groups "C1-C2", "C1-C3", "C2-C4", "C3-C4", "C3-C5", and "C4-C5" of two places, which are C1 "A post office", C2 "B station", C3 "C citizens square", C4 "D hospital branch", and C5 "cargo terminal", from coordinate information of each of the places. In addition, the relationship information acquisition section E31 calculates an absolute distance (here, meters) of a difference in position information between places belonging to each of the groups "C1-C2", "C1-C3", "C2-C4", "C3-C4", "C3-C5", and "C4-C5" of two places from position information of each of the places. Also, the relationship information acquisition section E31 calculates distance relationship information (here, MPP) of each of the groups of two places by "absolute distance / relative distance" using the relative distance and the absolute distance (refer to Figure 33 ). Also, a table of distance relationship information of the result of the processing involved, Figure 34 is obtained.

[0496] Next, the corresponding information acquisition section E32 acquires Figure 34the distance relationship information are clustered into the groups of "0.45", "0.41", "0.33", "0.41", "0.37", "0.44" and the groups of "523", "470", "941", "809". Further, the correspondence information acquisition section E32 acquires the place identifiers "C1", "C2", "C3", "C4" corresponding to the groups of "0.45", "0.41", "0.33", "0.41", "0.37", "0.44" and the place identifier "C5" corresponding to the groups of "523", "470", "941", "809".

[0497] Further, the correspondence information acquisition section E32 selects the place identifiers of the groups whose number of place identifiers is more than a predetermined condition. Here, the predetermined condition is, for example, "the number of place identifiers is the largest", "the number of place identifiers is equal to or more than a threshold value (for example, 3)", "the number of place identifiers is more than a threshold value (for example, 5)", and the like. Further, the correspondence information acquisition section E32 acquires the place identifiers "C1", "C2", "C3", "C4". That is, the correspondence information acquisition section E32 selects "A post office", "B station", "C citizens square", and "D hospital branch", and excludes (deletes) "cargo wharf". By the above processing, a table of the distance relationship information of Figure 35 That is, as a result of the checking processing using the distance relationship information, "cargo wharf" is excluded.

[0498] Next, the relationship information acquisition section E31 performs checking processing using the angle relationship information as follows. That is, the relationship information acquisition section E31 acquires the relative angle and the absolute angle between the places of each group belonging to two places of the groups "C1-C2", "C1-C3", "C2-C4", "C3-C4" in each place of C1 "A post office", C2 "B station", C3 "C citizens square", and C4 "D hospital branch". Further, the relationship information acquisition section E31 calculates the relative angle of the two places constituting the group using the coordinate information of the two places. In addition, the relationship information acquisition section E31 calculates the absolute angle of the two places constituting the group using the position information of the two places. In addition, Figure 36 3601 of FIG. 36 is an image of the relative angle (the angle with the bottom edge in the input image of the rectangle as a reference) between C1 and C3. In addition, Figure 36 3602 of FIG. 36 is an image of the absolute angle (the angle with the east-west direction in the real world as a reference) between C1 and C3. Further, the relationship information acquisition section E31 calculates the angle relationship information (Rotation) using the relative angle and the absolute angle. Here, the relationship information acquisition section E31 calculates the angle relationship information according to "angle relationship information = relative angle - absolute angle".

[0499] According to the above processing, the relationship information acquisition section E31 obtains the information shown in Figure 37 . In addition, the information shown in Figure 37 is made into a table of information shown in Figure 38 . In addition, the relationship information acquisition section E31 calculates (cos (angle relationship information), sin (angle relationship information)) from the angle relationship information, and calculates vector data in units of groups of two places (refer to Figure 38 ). Furthermore, the vector data can also be considered as the angle relationship information. Figure 39

[0500] Next, the corresponding information acquisition section E32 clusters the vectors of Figure 39 , and obtains the group 1 " (cos (-2), sin (-2)), (cos (-1), sin (-2)) " and the group 2 " (cos (-40), sin (-40)), (cos (-24), sin (-24)), (cos (-32), sin (-32)), (cos (-26), sin (-26)) ". Furthermore, the corresponding information acquisition section E32 obtains the place identifier "C2" corresponding to the group 1 and the place identifiers "C1", "C3", "C4" corresponding to the group 2.

[0501] Furthermore, the corresponding information acquisition section E32 selects the place identifiers of the group whose number of place identifiers satisfies a predetermined condition. Furthermore, it is assumed that the corresponding information acquisition section E32 obtains the place identifiers "C1", "C3", "C4". That is, the corresponding information acquisition section E32 selects "A post office", "C citizen square", "D hospital branch", and excludes (deletes) "B station".

[0502] Furthermore, as shown in Figure 40 , the corresponding information acquisition section E32 can also make links between the places corresponding to the appropriate values of the group 2 " (cos (-40), sin (-40)), (cos (-24), sin (-24)), (cos (-32), sin (-32)), (cos (-26), sin (-26)) ", for example, and obtain the place identifiers "C1", "C3", "C4" of the largest closed space. In addition, the corresponding information acquisition section E32 can also make links between the places corresponding to the appropriate values of the group 2 and delete the place identifier "C2" where the closed space does not exist, for example.

[0503] ​According to the above processing, the correspondence information acquisition section E32 selects, as the locations of the accumulated correspondence information of the object, the three locations of "A post office", "C civic center", and "D hospital branch". Further, the correspondence information acquisition section E32 acquires the correspondence information (coordinate information, position information) of the three respective locations. Further, the correspondence information can also be linked with a character string (the name of the location).

[0504] Next, the correspondence output section E41 accumulates the acquired three correspondence information. Further, the accumulation destination of the correspondence information is, for example, the correspondence information storage section C14, or an external device not shown.

[0505] Next, the correspondence information acquisition section E32 obtains the MPPs "0.41", "0.33", "0.44" corresponding to the three locations "C1", "C3", "C4". Next, the correspondence information acquisition section E32 calculates the representative value (here, the average value) of the three MPPs, and obtains "0.39" (refer to Figure 41 ).

[0506] Next, the scale acquisition section E33 acquires the scale information (for example, "0.39") using the representative value (MPP "0.39") of the acquired distance relationship information, and accumulates in association with the input image.

[0507] Next, the azimuth information acquisition section E39 performs the azimuth information acquisition processing as follows. That is, the azimuth information acquisition section E39 acquires the angle relationship information "-40" "-24" "-26" of each group among the location identifiers "C1" "C3" "C4". Next, the azimuth information acquisition section E39 acquires the representative value (here, the average value) of the angle relationship information "-30". Next, the azimuth information acquisition section E39 acquires "30 degrees" according to the calculation formula "azimuth information = -1 x (representative value of the angle relationship information)". Further, the azimuth information acquisition section E39 acquires the pattern (here, the angle information) indicating the azimuth inclined 30 degrees to the east side (30 degrees to the clockwise direction) from the true north. Further, the azimuth information output section E44 appends the pattern (the angle information) indicating the azimuth to the input image and performs output. An example of such output is Figure 42 . Figure 42 4201 of FIG. 20 is an example of the angle information.

[0508] Next, the region determination information acquisition section E34 performs the region determination processing as follows. For example, first, the region determination information acquisition section E34 detects each side (30a, 30b, 30c, 30d) of the region of the map of the input image by a contour extraction technique. Figure 30 Further, the technique of detecting the region by the contour extraction technique is a publicly known technique.

[0509] Next, the region determination information acquisition section E34 acquires the relative distance (here, the number of pixels) "720" from the location of the place identifier "Cl" to the right edge 30a. Further, the relative distance is the number of pixels of the vertical line from the location of "Cl" to the right edge 30a. Next, the region determination information acquisition section E34 calculates the absolute distance "280 m" to the right edge 30a from the operation "720 x 0.39" using the representative value (0.39) of the MPP and the number of pixels "720". Next, the region determination information acquisition section E34 acquires the position information [35.5918, 139.6053] of the right of the location of "Cl" in the right edge 30a using the position information of "Cl" and the absolute distance "280 m" (refer to Figure 43 ).

[0510] Similarly, the region determination information acquisition section E34 acquires the number of pixels "1199" from the location of the place identifier "C3" to the right edge 30a. Further, the region determination information acquisition section E34 multiplies the number of pixels "1199" by the representative value (0.39) of the MPP to calculate the absolute distance "468 m" from the location of the place identifier "C3" to the right edge 30a. Further, the region determination information acquisition section E34 acquires the position information [35.5849, 139.6096] of the right of the location of "C3" in the right edge 30a using the position information of "C3" and the absolute distance "468 m" (refer to Figure 43 ). Further, similarly, the region determination information acquisition section E34 acquires the position information of the right of the location of "C4" in the right edge 30a.

[0511] Next, the region determination information acquisition section E34 acquires the information (the information of the straight line) of the right edge of each of the locations of "Cl", "C3", and "C4" using the position information on the right edge corresponding to each of the locations of "Cl", "C3", and "C4" and the azimuth information. Further, the relevant conceptual diagram is Figure 44 . Next, the region determination information acquisition section E34 acquires the representative value of the right edge of the three right edge information (here, the median 44c). The representative value of the right edge is determined as the boundary line of the right edge in the accurate map information. Further, the representative value of the right edge can be the average value.

[0512] The region determination information acquisition section E34 performs the same processing as the processing of acquiring the boundary line of the right edge in the accurate map information to acquire the information of the boundary line of the lower edge (45a), the left edge (45b), and the upper edge (45c) in the accurate map information. Next, the region determination information acquisition section E34 obtains the intersection points (XI, Yl), (X2, Y2), (X3, Y3), and (X4, Y4) of the four edges (44c, 45a, 45b, and 45c) (refer to Figure 45). In addition, the intersection is (latitude, longitude).

[0513] Through the above processing, the region determination information acquisition section E34 acquires region determination information of a region that determines accurate map information corresponding to a region of a map in an input image. In addition, the region determination information is, for example, information of four intersection points, or information of two points (X1, Y1), (X3, Y3), or information of two points (X2, Y2), (X4, Y4).

[0514] Next, the processing section E3 reads accurate map information including a region determined by the region determination information. Next, the processing section E3 constructs accurate map information that explicitly indicates the region determined by the region determination information in the accurate map information. Next, the output section E4 outputs the accurate map information that explicitly indicates the region determined by the region determination information. An output example of such accurate map information is Figure 46 . In addition, the region is a rectangle, and is 4601.

[0515] Next, by processing based on the flowchart of Figure 27 , the additional place acquisition section E35 searches a place dictionary and acquires one or more pieces of place position information that enter a region determined by the region determination information. Next, the additional place acquisition section E35 acquires place information possessed by each of the one or more pieces of place position information. Next, the additional coordinate acquisition section E37 acquires coordinate information corresponding to the position information possessed by each of the one or more pieces of place position information. Next, the place recording section E38 configures a string of the acquired place information on the input image at a position indicated by the acquired coordinate information with respect to each of the place position information. According to the above processing, output target accurate map information is constructed.

[0516] Next, the output section E4 outputs the constructed accurate map information. According to the present embodiment, above, it is possible to automatically acquire appropriate correspondence information that corresponds and associates coordinate information of a place name or the like on map representation data with position information.

[0517] In addition, according to the present embodiment, it is possible to acquire a range in the real world supported by map representation data.

[0518] In addition, according to the present embodiment, it is possible to attach new place information to map representation data.

[0519] Further, according to the present embodiment, it is possible to acquire orientation information of map representation data.

[0520] In addition, the map representation data processing apparatus D explained in the third embodiment can use one or more pieces of correspondence information automatically acquired in the present embodiment to explicitly indicate a current position on map representation data.

[0521] In this case, the current position can be indicated on the map expression data using two or more pieces of correspondence information automatically acquired by the map expression data processing apparatus E.

[0522] Further, the processing in the present embodiment can also be realized by software. Also, the software can be distributed by software download or the like. In addition, the software can be distributed by being recorded on a recording medium such as a CD-ROM. Furthermore, the same applies to the other embodiments in the present specification. Furthermore, the software that realizes the information processing apparatus in the present embodiment is a program such as the following. That is, the program is a program for causing a computer to function as a map expression data reception section that receives map expression data of a kind of a map that is any one or more of a pictorial map, a hand-drawn map, and an outline map; a character string acquisition section that acquires three or more character strings from the map expression data; a coordinate information acquisition section that acquires coordinate information in the map expression data as coordinate information corresponding to the three or more respective character strings acquired by the character string acquisition section; a position information acquisition section that acquires position information corresponding to the position information of each of the three or more respective character strings acquired by the character string acquisition section using a place dictionary having one or more pieces of place-position information that associates place information of a place with position information of a position of the place; a relationship information acquisition section that acquires relationship information related to a relationship between the correspondence information of the group of the coordinate information and the position information corresponding to the three or more respective character strings acquired by the character string acquisition section; a correspondence information acquisition section that acquires, using the relationship information, two or more pieces of correspondence information corresponding to the relationship information that satisfies a relationship decided in advance among the three or more pieces of correspondence information; and a correspondence output section that accumulates the two or more pieces of correspondence information acquired by the correspondence information acquisition section.

[0523] In addition, Figure 47 The appearance of a computer that executes the program described in the present specification, and realizes the map expression data processing apparatus C and the like of the above-described embodiments is shown. The above-described embodiments can be realized by computer hardware and a computer program executed thereon. Figure 47 is a diagram showing an overview of the computer system 300, Figure 48 is a block diagram of the system 300.

[0524] In Figure 47 , the computer system 300 includes a computer 301 including a CD-ROM drive, a keyboard 302, a mouse 303, and a monitor 304.

[0525] In Figure 48In the present embodiment, the computer 301 includes, in addition to the CD-ROM drive 3012, an MPU 3013, a bus 3014 connected to the CD-ROM drive 3012 and the like, a ROM 3015 for storing programs of boot programs and the like, a RAM 3016 connected to the MPU 3013, temporarily storing commands of application programs, and used for providing a temporary storage space, and a hard disk 3017 for storing application programs, system programs, and data. Here, although not shown, the computer 301 can further include a network card that provides a connection to a LAN.

[0526] A program that causes the computer system 300 to execute the functions of the map expression data processing apparatus C and the like of the above-described embodiments can also be stored in the CD-ROM 3101, and inserted into the CD-ROM drive 3012, and further transferred to the hard disk 3017. Alternatively, the program can be sent to the computer 301 through a network that is not shown, and stored in the hard disk 3017. The program is loaded into the RAM 3016 at the time of execution. The program can also be loaded directly from the CD-ROM 3101 or the network.

[0527] The program can not include an operating system (OS) that causes the computer 301 to execute the functions of the map expression data processing apparatus C and the like of the above-described embodiments, or a third-party program, and the like. The program can call an appropriate function (module) in a controlled manner, and include only a command portion that obtains a desired result. How the computer system 300 acts is known, and detailed explanation is omitted.

[0528] Further, in the above-described program, in the steps of transmitting information, and receiving information, and the like, a process performed by hardware, for example, a process performed by a modem and an interface card, and the like in the transmitting step (a process performed only by hardware) is not included.

[0529] In addition, the computer that executes the above-described program can be a single one, or a plurality of. That is, either a centralized processing can be performed, or a distributed processing can be performed.

[0530] In addition, in the above-described embodiments, two or more communication units that exist in one apparatus can be physically one.

[0531] In addition, in the above-described embodiments, each processing can be realized by either a centralized processing by a single apparatus, or a distributed processing by a plurality of apparatuses. That is, the map expression data processing apparatus C can also act as a stand alone.

[0532] The present application is not limited to the above-described embodiments, and various modifications can be made, and these are, of course, included in the scope of the present application.

[0533] Industrial applicability

[0534] As above, the map expression data processing apparatus to which the present application relates has the effect that coordinate information of a place name or the like on map expression data and position information can be automatically acquired in association with each other, and is useful as a map expression data processing apparatus or the like.

[0535] Explanation of reference numerals

[0536] A: learning apparatus

[0537] B: classification apparatus

[0538] C, D, E: map expression data processing apparatus

[0539] A1, B1, C1, D1: storage section

[0540] A2, B2, C2, D2: reception section

[0541] A3, B3, C3, D3, E3: processing section

[0542] A4, B4, C4, D4, E4: output section

[0543] A11: learning source information storage section

[0544] A12: learner storage section

[0545] A31: learning section

[0546] A41: accumulation section

[0547] B21: character string reception section

[0548] B31: classification section

[0549] B41: label output section

[0550] C11: place dictionary storage section

[0551] C12: map expression data storage section

[0552] C13: accurate map information storage section

[0553] C14: correspondence information storage section

[0554] C21: map expression data reception section

[0555] C31: character string acquisition section

[0556] C32: classification section

[0557] C33: grouping section

[0558] C34: feature location detection section

[0559] C35: coordinate information acquisition section

[0560] C36: position information acquisition section

[0561] C37: region determination information acquisition section

[0562] C38: orientation information acquisition section

[0563] C41: correspondence output section

[0564] C42: accurate map information output section

[0565] C43: orientation information output section

[0566] C371: first region determination information acquisition unit

[0567] C372: outside string decision unit

[0568] C373: size information acquisition unit

[0569] C374: distance information acquisition unit

[0570] C375: second region determination information acquisition unit

[0571] D31: current position information acquisition section

[0572] D32: coordinate information acquisition section

[0573] D33: data construction section

[0574] D41: map representation data output section

[0575] E31: relationship information acquisition section

[0576] E32: correspondence information acquisition section

[0577] E33: scale acquisition section

[0578] E34: region determination information acquisition section

[0579] E35: additional place acquisition section

[0580] E36: additional position acquisition section

[0581] E37: additional coordinate acquisition section

[0582] E38: place note section

[0583] E39: orientation information acquisition section

[0584] E41: correspondence output section

[0585] E42: scale information output section

[0586] E43: region determination information output section

[0587] E44: orientation information output section

[0588] E321: judging unit

[0589] E322: corresponding information obtaining unit

Claims

1. A map expression data processing apparatus comprising: a map expression data receiving section that receives map expression data of a map of a kind of any one or more of a pictorial map, a hand-drawn map, and an outline map; a character string acquisition section that acquires a character string from the map expression data; a coordinate information acquisition section that acquires coordinate information that is information that specifies a coordinate position corresponding to the character string acquired by the character string acquisition section and is information that specifies a relative coordinate position in the map expression data; a position information acquisition section that acquires position information corresponding to place information that is a character string acquired by the character string acquisition section, using a place dictionary having one or more pieces of place position information that is information that associates place information that specifies a place and position information that specifies a position of the place; and a correspondence output section that outputs the coordinate information acquired by the coordinate information acquisition section and the position information acquired by the position information acquisition section in association with each other. The character string acquisition section acquires two or more character strings from the map expression data. The apparatus further comprises a classification section that determines whether each of the two or more character strings acquired by the character string acquisition section is a character string that specifies a place. The correspondence output section outputs coordinate information and position information corresponding to a character string that is determined by the classification section to be a character string that specifies a place in association with each other. The apparatus further comprises a learner storage section that imparts two or more pieces of learner source information having a character string having two or more characters and a label that is a label related to a number of occurrences in a place dictionary and is any one of two or more labels to a learning module of machine learning, executes the learning module, and stores a learner acquired thereby. The classification section classifies each of the two or more character strings acquired by the character string acquisition section into one of place information and non-place information using the learner by a technique of machine learning. The correspondence output section outputs coordinate information and position information corresponding to a character string classified by the classification section as place information in association with each other.

2. The map expression data processing apparatus according to claim 1, wherein the apparatus further comprises a feature place detection section that detects a feature place that is a place represented by a feature pattern that is a characteristic pattern determined in advance from the map expression data, The coordinate information acquisition section acquires coordinate information corresponding to the feature pattern. The position information acquisition section acquires position information corresponding to place information that is a character string corresponding to the feature pattern using the place dictionary.

3. The map expression data processing apparatus according to claim 1, wherein the character string acquisition section acquires three or more character strings from the map expression data. ​ ​ ​ ​ ​ ​ ​ ​ Further provided is a grouping section that determines two or more of the three or more pieces of position information acquired by the position information acquisition section, the two or more pieces of position information indicating positions close to satisfying a predetermined condition, The corresponding output section outputs each of the two or more pieces of position information determined by the grouping section in association with each of the two or more pieces of coordinate information acquired by the coordinate information acquisition section.

4. The map expression data processing apparatus according to claim 3, further comprising: an accurate map information storage section that stores accurate map information as accurate map information; a region determination information acquisition section that acquires region determination information that determines a region in the accurate map information as a region containing positions indicated by the two or more pieces of position information determined by the grouping section; and an accurate map information output section that outputs the accurate map information in a manner that is aware of the region determined by the region determination information.

5. The map expression data processing apparatus according to claim 4, wherein: the region determination information acquisition section includes: a first region determination information acquisition unit that acquires first region determination information that determines a first region in the accurate map information as a region containing positions indicated by the two or more pieces of position information determined by the grouping section; an outside string determination unit that determines a string corresponding to one or more pieces of position information located at the outermost side among the pieces of position information acquired by the position information acquisition section; a size information acquisition unit that acquires size information related to a size from an outside of the one or more strings determined by the outside string determination unit to a corner of the map expression data; a distance information acquisition unit that acquires distance information that determines a distance corresponding to the size information acquired by the size information acquisition unit using two or more groups of the coordinate information acquired by the coordinate information acquisition section and the position information acquired by the position information acquisition section; and a second region determination information acquisition unit that acquires second region determination information that determines a second region as a region that expands the first region determined by the first region determination information using the distance information, and the accurate map information output section outputs the accurate map information in a manner that is aware of the region determined by the second region determination information.

6. The map expression data processing apparatus according to claim 5, wherein: the region determination information or the second region determination information includes information of two points that constitute a rectangle, and further comprising: an orientation information acquisition section that acquires orientation information related to an orientation of the map expression data using the region determination information or the second region determination information; and an orientation information output section that outputs the orientation information.

7. The map expression data processing apparatus according to claim 1, further comprising: ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ a relationship information acquisition section that acquires relationship information related to a relationship between correspondence information that corresponds to a group of the coordinate information and the position information of each of three or more strings acquired by the string acquisition section; a correspondence information acquisition section that acquires, using the relationship information, two or more pieces of correspondence information that correspond to relationship information that satisfies a predetermined relationship among the three or more pieces of correspondence information; and a correspondence output section that stores the two or more pieces of correspondence information acquired by the correspondence information acquisition section.

8. The map expression data processing apparatus according to claim 7, wherein the relationship information acquisition section acquires, in units of a group of two strings, relative relationship information between correspondence information that corresponds to two respective strings among the three or more strings acquired by the string acquisition section, the correspondence information acquisition section includes: a determination unit that groups the three or more pieces of relationship information acquired by the relationship information acquisition section and determines relationship information that does not enter any group; and a correspondence information acquisition unit that excludes, from correspondence information that corresponds to the relationship information that does not enter the group, correspondence information that corresponds to the relationship information that does not enter the group and acquires only the remaining two or more pieces of correspondence information.

9. The map expression data processing apparatus according to claim 7, wherein the relationship information acquisition section acquires, in units of a group of the two strings, distance relationship information that is relationship information indicating a relationship between a relative distance that is a difference in coordinate information of the two strings and an absolute distance that is a difference in position information of the two strings, the correspondence information acquisition section determines whether three or more respective pieces of the distance relationship information satisfy a predetermined relationship, excludes, from correspondence information that corresponds to the distance relationship information that does not satisfy the predetermined relationship, correspondence information that corresponds to the distance relationship information that does not satisfy the predetermined relationship, and acquires only the remaining two or more pieces of correspondence information.

10. The map expression data processing apparatus according to claim 9, further comprising: a scale acquisition section that acquires a representative value of the distance relationship information from the three or more pieces of the distance relationship information and acquires scale information of map expression data using the representative value of the distance relationship information; and a scale information output section that outputs the scale information.

11. The map expression data processing apparatus according to claim 10, further comprising: a region determination information acquisition section that acquires a relative perpendicular distance that is a perpendicular distance of coordinate information that corresponds to a string acquired by the string acquisition section from each of four sides of the map expression data and acquires region determination information that determines a range of the map expression data in the real world using the relative perpendicular distance and the scale information; and a region determination information output section that outputs the region determination information.

12. The map expression data processing apparatus according to claim 11, further comprising: an additional place acquisition unit that acquires place information of a range indicated by the region determination information from the place dictionary; an additional position acquisition unit that acquires position information corresponding to the place information from the place dictionary; an additional coordinate acquisition unit that acquires coordinate information corresponding to the position information; and a place recording unit that arranges the place information at a position on the map expression data and a position indicated by the coordinate information.

13. The map expression data processing apparatus according to claim 7, wherein the relationship information acquisition unit acquires angle relationship information, which is relationship information indicating a relationship between a relative angle that is an angle obtained from coordinate information of the two strings and an absolute angle that is an angle obtained from position information of the two strings, in units of groups of the two strings, the correspondence information acquisition unit judges whether each of the three or more angle relationship information satisfies a predetermined relationship, excludes correspondence information corresponding to angle relationship information that does not satisfy the predetermined relationship from correspondence information corresponding to the three or more strings acquired by the string acquisition unit, and acquires only two or more correspondence information remaining.

14. The map expression data processing apparatus according to claim 13, further comprising: an orientation information acquisition unit that acquires a representative value of the angle relationship information from the three or more angle relationship information, and acquires orientation information indicating an orientation of map expression data using the representative value of the angle relationship information; and an orientation information output unit that outputs the orientation information.

15. A map expression data processing apparatus comprising: a map expression data storage unit that stores map expression data; a correspondence information storage unit that stores two or more correspondence information that is a group of coordinate information and position information output by the map expression data processing apparatus according to any one of claims 1 to 14; a current position information acquisition unit that acquires current position information that determines a current position; a coordinate information acquisition unit that acquires coordinate information corresponding to the current position information using the two or more correspondence information; a data construction unit that constructs, on the map expression data, current position additional map expression data that is map expression data explicitly indicating a position indicated by the coordinate information; and a map expression data output unit that outputs the current position additional map expression data.

16. A method of generating correspondence information, implemented by a map expression data reception unit, a string acquisition unit, a coordinate information acquisition unit, a position information acquisition unit, and a correspondence output unit, the method comprising: a map expression data reception step of receiving, by the map expression data reception unit, map expression data that is a kind of map of any one or more of a pictorial map, a hand-drawn map, and an outline map; a string acquisition step of acquiring, by the string acquisition unit, a string from the map expression data; a coordinate information acquisition step of acquiring, by the coordinate information acquisition unit, coordinate information of the string; a position information acquisition step of acquiring, by the position information acquisition unit, position information of the string; and a correspondence output step of outputting, by the correspondence output unit, correspondence information that is a group of the coordinate information and the position information. The coordinate information obtaining step obtains coordinate information from the coordinate information obtaining section, the coordinate information being information that specifies a coordinate position corresponding to a character string obtained in the character string obtaining step, and being information that specifies a relative coordinate position in the map expression data; The position information obtaining step obtains position information from the position information obtaining section using a place dictionary having one or more pieces of place position information, the place position information being information that associates place information that specifies a place with position information that specifies a position of the place, the position information being position information corresponding to place information obtained as a character string in the character string obtaining step; and The correspondence output step outputs coordinate information obtained in the coordinate information obtaining step in association with position information obtained in the position information obtaining step, In the character string obtaining step, two or more character strings are obtained from the map expression data, The method of generating correspondence information further includes a classification step of determining whether each of the two or more character strings obtained in the character string obtaining step is a character string that specifies a place, In the correspondence output step, coordinate information and position information corresponding to a character string determined to be a character string that specifies a place in the classification step are output in association with each other, The map expression data processing apparatus further includes a learner storage step of giving a learner module that performs machine learning two or more pieces of learning source information having a character string and a label, the character string having two or more characters, the label being a label related to a number of occurrences in a place dictionary, and being any of two or more labels, executing the learner module, and storing the obtained learner, In the classification step, each of the two or more character strings obtained in the character string obtaining step is classified into one of place information and non-place information by a technique of machine learning using the learner, In the correspondence output step, coordinate information and position information corresponding to a character string classified as place information in the classification step are output in association with each other.

17. A recording medium storing a program that causes a computer capable of accessing a place dictionary storage section that stores a place dictionary having one or more pieces of place position information that are information that associates place information that specifies a place with position information that specifies a position of the place to function as each of the following sections: a map expression data receiving section that receives map expression data of a map of a kind that is any one or more of a diagrammatic map, a hand-drawn map, and an outline map; a character string obtaining section that obtains a character string from the map expression data; a coordinate information obtaining section that obtains coordinate information, the coordinate information being information that specifies a coordinate position corresponding to a character string obtained by the character string obtaining section, and being information that specifies a relative coordinate position in the map expression data; The position information acquisition section acquires position information corresponding to the place information of the string acquired by the string acquisition section using the place dictionary. And The corresponding output section outputs the coordinate information acquired by the coordinate information acquisition section and the position information acquired by the position information acquisition section in association with each other in correspondence with the string, The string acquisition section acquires two or more strings from the map expression data, The program further causes the computer to function as a classification section that determines whether each of the two or more strings acquired by the string acquisition section is a string of a certain place, The corresponding output section outputs the coordinate information and the position information in association with each other in correspondence with the string determined by the classification section to be a string of a certain place, The program further causes the computer to function as a learner storage section that imparts two or more pieces of learning source information having a string and a label to a learning module of machine learning, executes the learning module, and stores the acquired learner, the string having two or more characters, the label being a label related to the number of occurrences in a place dictionary and being any of two or more labels, The classification section uses each of the two or more strings acquired by the string acquisition section and the learner to perform classification processing of each of the two or more strings into one of place information and non-place information by a technique of machine learning, The corresponding output section outputs the coordinate information and the position information in association with each other in correspondence with the string classified as place information by the classification section.

Images