WO2020262312A1 - 地図表現データ処理装置、対応情報の生産方法、およびプログラム - Google Patents
地図表現データ処理装置、対応情報の生産方法、およびプログラム Download PDFInfo
- Publication number
- WO2020262312A1 WO2020262312A1 PCT/JP2020/024435 JP2020024435W WO2020262312A1 WO 2020262312 A1 WO2020262312 A1 WO 2020262312A1 JP 2020024435 W JP2020024435 W JP 2020024435W WO 2020262312 A1 WO2020262312 A1 WO 2020262312A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- information
- acquisition unit
- representation data
- unit
- map
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/53—Querying
- G06F16/532—Query formulation, e.g. graphical querying
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/55—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/587—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using geographical or spatial information, e.g. location
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/907—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/909—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using geographical or spatial information, e.g. location
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/19—Recognition using electronic means
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
- G09B29/00—Maps; Plans; Charts; Diagrams, e.g. route diagram
- G09B29/10—Map spot or coordinate position indicators; Map reading aids
Definitions
- the present invention relates to a map representation data processing device or the like that automatically acquires and outputs correspondence information indicating the correspondence between a point name acquired from map representation data and location information.
- the position information acquisition unit that acquires the position information corresponding to the point information that is the character string acquired by the character string acquisition unit using the point dictionary, and the coordinate information and the position information acquisition unit acquired by the coordinate information acquisition unit. It is a map representation data processing device including a corresponding output unit that outputs in association with the generated position information.
- coordinate information such as place names on map representation data can be automatically acquired in association with position information.
- the character string acquisition unit acquires two or more character strings from the map representation data, and the character string acquisition unit acquires 2
- the classification unit for determining whether or not each of the above character strings is a character string for specifying a point is further provided, and the corresponding output unit corresponds to the character string determined by the classification unit to be a character string for specifying a point. It is a map representation data processing device that outputs the coordinate information to be output in association with the position information.
- the map representation data processing device of the third invention is a character string having two or more characters and a label relating to the number of appearances in the point dictionary for the second invention, and is two or three or more types. Two or more learning source information having one of the labels is given to the learning module of machine learning, the learning module is executed, and a learning device storage unit for storing the acquired learning device is further provided.
- the classification unit uses the two or more character strings acquired by the character string acquisition unit and the learner to obtain either point information or non-point information for each of the two or more character strings by a machine learning technique.
- the corresponding output unit is a map representation data processing device that performs the deciding classification process and outputs the coordinate information corresponding to the character string classified into the point information by the classification unit in association with the position information.
- a characteristic symbol which is a predetermined characteristic symbol is expressed from the map representation data. It further includes a feature location detection unit that detects the feature location, the coordinate information acquisition unit acquires the coordinate information corresponding to the feature symbol, and the position information acquisition unit uses the point dictionary to make the feature symbol. It is a map representation data processing device that acquires the position information corresponding to the point information which is the corresponding character string.
- the graphic feature points on the map representation data can be acquired, and the coordinate information and the position information of the graphic feature points can be associated and automatically acquired.
- the map representation data processing device of the fifth invention is a map representation data processing device whose characteristic symbol is one or more kinds of symbols of an intersection or a bridge, as opposed to the fourth invention.
- the character string acquisition unit acquires three or more character strings from the map representation data, and the position information.
- the grouping unit further includes a grouping unit that determines two or more position information indicating positions that are close enough to satisfy a predetermined condition, and the corresponding output unit is determined by the grouping unit.
- This is a map representation data processing device that outputs the two or more position information obtained in association with each of the two or more coordinate information acquired by the coordinate information acquisition unit.
- a character string such as an appropriate place name on the map representation data can be extracted, and the coordinate information such as the place name and the position information can be associated and automatically acquired.
- the accurate map information storage unit for storing accurate map information, which is accurate map information, and the grouping unit have been determined for the sixth invention. It is an area including the position indicated by the above position information, and the accurate map is in a manner in which the area specific information acquisition unit for acquiring the area specific information for specifying the area in the accurate map information and the area specified by the area specific information can be known. It is a map representation data processing device further provided with an accurate map information output unit that outputs information.
- the range of accurate map information corresponding to the map representation data can be clarified.
- the area specific information acquisition unit is an area including a position indicated by two or more position information determined by the grouping unit.
- the first area specific information acquisition means for acquiring the first area specific information for specifying the first area which is the area in the accurate map information, and the outermost position information acquired by the position information acquisition unit 1
- the outer character string determining means for determining the character string corresponding to each of the above position information and the size information regarding the size to the corner of the map representation data, which is outside the one or more character strings determined by the outer character string determining means.
- the distance corresponding to the size information acquired by the size information acquisition means by using two or more sets of the size information acquisition means to be acquired, the coordinate information acquired by the coordinate information acquisition unit, and the position information acquired by the position information acquisition unit.
- the second area specific information for specifying the second area which is an area obtained by expanding the first area specified by the first area specific information, is acquired.
- the accurate map information output unit is a map representation data processing device that outputs accurate map information in a manner in which the area specified by the second area specific information can be known.
- the area identification information or the second area identification information includes information of two points constituting the rectangle, and the area identification It is a map representation data processing device further including a direction information acquisition unit that acquires direction information regarding the direction of map representation data using information or second area specific information, and a direction information output unit that outputs direction information.
- the map representation data processing device of the tenth invention has coordinate information and position information corresponding to each of three or more character strings acquired by the character string acquisition unit for any one of the first to fourth inventions.
- the relationship information acquisition unit that acquires the relationship information related to the relationship between the corresponding information, which is a set with, and the relationship information are used to correspond to the relationship information that satisfies the predetermined relationship among the three or more correspondence information.
- the related information acquisition unit is a character string of two of the three or more character strings acquired by the character string acquisition unit.
- the relative relationship information between the corresponding information corresponding to is acquired for each set of two character strings, and the corresponding information acquisition unit groups the three or more relationship information acquired by the relationship information acquisition unit into a group. From the judgment means for determining the relational information that does not enter and the correspondence information corresponding to each of the three or more character strings acquired by the character string acquisition unit, the correspondence information corresponding to the relational information that does not enter the group is excluded, and the remaining 2
- the relational information acquisition unit is for each set of two character strings, and the coordinate information of the two character strings.
- the distance relationship information which is the relationship information indicating the relationship between the relative distance, which is the difference between the two character strings, and the absolute distance, which is the difference between the position information of the two character strings, is acquired, and the correspondence information acquisition unit has three or more distance relationship information. It is determined whether or not the predetermined relationship is satisfied, and the distance relationship information that does not satisfy the predetermined relationship is supported from the corresponding information corresponding to each of the three or more character strings acquired by the character string acquisition unit. It is a map representation data processing device that acquires only the remaining two or more correspondence information excluding the correspondence information.
- the map representation data processing device of the thirteenth invention acquires the representative value of the distance-related information from three or more distance-related information for the twelfth invention, and uses the representative value of the distance-related information.
- the map representation data processing device further includes a scale acquisition unit for acquiring scale information of map representation data and a scale information output unit for outputting scale information.
- the map representation data processing device of the fourteenth invention has the coordinate information corresponding to the character string acquired by the character string acquisition unit and each of the four sides of the map representation data for the thirteenth invention.
- the area-specific information acquisition unit that acquires the relative vertical distance, which is the vertical distance to the side, and acquires the area-specific information that specifies the range of the map representation data in the real world using the relative vertical distance and the scale information, and the area. It is a map representation data processing device further including an area specific information output unit for outputting specific information.
- the range in the real world supported by the map representation data can be acquired.
- the map representation data processing device of the fifteenth invention corresponds to the fourteenth invention with an additional point acquisition unit that acquires the point information in the range indicated by the area specific information from the point dictionary and the point information.
- An additional position acquisition unit that acquires the position information to be performed from the point dictionary, an additional coordinate acquisition unit that acquires the coordinate information corresponding to the position information, and a position on the map representation data, and the point information is displayed at the position indicated by the coordinate information.
- the relation information acquisition unit is for each set of two character strings, and the coordinate information of the two character strings.
- the angle relationship information which is the relationship information indicating the relationship between the relative angle, which is the angle obtained from, and the absolute angle, which is the angle obtained from the position information of the two character strings, is acquired, and the corresponding information acquisition unit is each of 3 or more. It is determined whether or not the angle relationship information satisfies the predetermined relationship, and the angle relationship that does not satisfy the predetermined relationship is selected from the corresponding information corresponding to each of the three or more character strings acquired by the character string acquisition unit. It is a map representation data processing device that acquires only the remaining two or more correspondence information excluding the correspondence information corresponding to the information.
- the map representation data processing apparatus of the seventeenth invention acquires the representative value of the angle-related information from three or more angle-related information for the sixteenth invention, and uses the representative value of the angle-related information.
- the map representation data processing device further includes a direction information acquisition unit that acquires direction information indicating the direction of the map representation data, and a direction information output unit that outputs the direction information.
- the direction information of the map representation data can be acquired.
- the map representation data processing device of the eighteenth invention has a map representation data storage unit for storing map representation data and coordinate information output by any one of the first to seventeenth map representation data processing devices.
- a correspondence information storage unit that stores two or more correspondence information that is a set of the position information acquired by the position information acquisition unit, a current position information acquisition unit that acquires the current position information that specifies the current position, and two or more.
- the coordinate information acquisition unit that acquires the coordinate information corresponding to the current position information using the corresponding information of, and the current location additional map expression data that is the map expression data in which the position indicated by the coordinate information is clearly indicated on the map expression data.
- It is a map representation data processing apparatus including a data configuration unit constituting the above and a map representation data output unit for outputting current location additional map representation data.
- the current location can be shown on the map representation data using the automatically acquired two or more correspondence information.
- coordinate information such as place names on map representation data can be automatically acquired in association with position information.
- Block diagram of learning device A according to the first embodiment Block diagram of sorting device B A flowchart for explaining an operation example of the learning device A
- Explanatory drawing of a concrete operation example of the learning device A and the classification device B Diagram showing an example of the learning source information
- Block diagram of the map representation data processing device C according to the second embodiment A flowchart for explaining an operation example of the map representation data processing device C.
- Flow chart explaining an example of the grouping process Flow chart explaining an example of the same area identification process
- Flowchart explaining an example of the same direction acquisition process Flow chart explaining an example of the output processing Diagram showing examples of the map representation data, etc. Diagram showing examples of the map representation data, etc.
- FIG. 1 Flow chart explaining an example of the same area identification process Flowchart explaining an example of the same direction information acquisition process Flowchart explaining an example of the same point addition process Flowchart explaining an example of the output map configuration processing
- the figure which shows the specific processing example of the map representation data processing apparatus E Diagram showing an example of the map representation data
- the learning source information is the information to be learned.
- the learning source information has a character string and a label.
- the character string has two or more characters.
- the label is information on the number of occurrences in the point dictionary described later.
- the label can take information on any one of two or more types of labels.
- the label is the first label indicating that the number of appearances in the point dictionary is one, the second label indicating that the number of appearances in the point dictionary is 0, and the number of appearances in the point dictionary is two or more. It is preferable that it is one of the third labels indicating that it is present. That is, it is preferable that the label can take any one of the three pieces of information.
- the point dictionary is a dictionary related to point names.
- the point dictionary has, for example, two or more corresponding information having a point name and position information for specifying the position of the place specified by the point name.
- a point name is, for example, a place name, a landscape name, a specific point name, a prefecture name, a municipality name, a river, a mountain, a park, a scenic spot, or the like.
- the point dictionary has, for example, two or more corresponding information having the point name and the information of the description of the point specified by the point name.
- the learning device is stored in the learning device storage unit A12.
- the learner can be called a classifier. Further, the learner may be called a classification model, a prediction model, or the like.
- a learner is information acquired by a learning module that employs machine learning techniques.
- the machine learning technique may be deep learning, SVM, decision tree, random forest, or the like. It is preferable that the machine learning technique is, for example, a technique using a recurrent neural network (RNN). That is, it is preferable that the learner is an RNN.
- RNN recurrent neural network
- Reception department A2 receives various information and instructions.
- the acceptance means acceptance of user input, reception from an external device, and the like. However, it suffices if various information and instructions can be obtained by acceptance.
- the input means for various information and instructions may be any, for example, a touch panel, a keyboard, a mouse, or a menu screen.
- the processing unit A3 performs various processes.
- the various processes are, for example, processes performed by the learning unit A31.
- the learning unit A31 uses two or more learning source information of the learning source information storage unit A11 to perform learning processing by a machine learning technique, and acquires a learning device.
- the learning unit A31 gives each character constituting the character string of the two or more learning source information of the learning source information storage unit A11 to the learning module of machine learning in order, and assigns a label to be paired with the character string.
- the process of sequentially giving each character constituting the character string to the machine learning learning module is, for example, a process of giving a stream of the character string to the machine learning learning module. That is, it can be said that the learning unit A31 learns the stream of the character string possessed by each of the two or more learning source information.
- the machine learning learning module is stored in, for example, the storage unit A1.
- a machine learning technique for acquiring a learning device for example, deep learning, SVR, random forest, decision tree, etc.
- the learning module of machine learning is, for example, a function in a machine learning framework such as fastText, tinySVM, TensorFlow, and various random forest functions.
- the learning device acquired by the learning unit A31 is information for outputting a label when a character string is input and given to a machine learning classification module.
- the learning device acquired by the learning unit A31 is information for outputting a label, for example, when characters constituting a character string are given one by one in order.
- the learning unit A31 may be considered to include the learning module or may not include the learning module.
- Output unit A4 outputs various information.
- the various information is, for example, a learning device.
- the output is usually storage on a recording medium, but display on a display, projection using a projector, printing by a printer, sound output, transmission to an external device, other processing devices, etc. It may be considered that the concept includes the delivery of the processing result to the program or the like.
- the points representing the map representation data are, for example, the center of gravity point of the map representation data, any point at the end of the map representation data, or any point constituting the boundary of the area of the map representation data.
- the map representation data is included in the file.
- the map representation data may be data in the database, and the data format and management method are not limited.
- the file may contain two or more map representation data.
- one map representation data may be realized by two or more files. That is, one map representation data may be divided into two or more files.
- the learning device stored in the learning device A is stored in the learning device storage unit A12 constituting the classification device B.
- Reception department B2 receives various information and instructions.
- the acceptance means for example, acceptance of user input, reception from an external device, and the like. However, it suffices if various information and instructions can be obtained by acceptance.
- the input means for various information and instructions may be any, for example, a touch panel, a keyboard, a mouse, or a menu screen.
- the character string reception unit B21 accepts a character string.
- the acceptance means, for example, acceptance of information input from an input device such as a keyboard, mouse, or touch panel, reception of information transmitted via a wired or wireless communication line, an optical disk, a magnetic disk, a semiconductor memory, or the like. It is a concept including acceptance of information read from the recording medium of.
- the character string reception unit B21 performs character recognition processing on the map information, for example, and acquires one or more character strings.
- the processing unit B3 performs various processes.
- the various processes are, for example, processes performed by the classification unit B31.
- the classification unit B31 acquires a label corresponding to the character string by a machine learning technique using the character string received by the reception unit B2 and the learner of the learner storage unit A12.
- the classification unit B31 gives each of two or more characters constituting the character string received by the reception unit B2 to the machine learning classification module in order for each character, and supplies the learner of the learner storage unit A12 to the classification module. To execute the classification module and get the label.
- classification unit B31 may be considered to include the classification module or may not include the classification module.
- Output unit B4 outputs various information.
- the various types of information are, for example, labels acquired by the classification unit B31.
- the label output unit B41 outputs the label acquired by the classification unit B31.
- the output means storage on a recording medium, display on a display, projection using a projector, printing by a printer, sound output, transmission to an external device, transmission to another processing device, another program, or the like. It may be considered that the concept includes the delivery of the processing result.
- the process of storing information in the storage unit A1 etc. does not matter.
- the information may be stored in the storage unit A1 or the like via the recording medium, or the information transmitted via the communication line or the like may be stored in the storage unit A1 or the like.
- the information input via the input device may be stored in the storage unit A1 or the like.
- the reception unit A2, the reception unit B2, and the character string reception unit B21 can be realized by a device driver of an input means such as a touch panel or a keyboard, a menu screen control software, or the like.
- the processing unit A3, the learning unit A31, the storage unit A41, the processing unit B3, and the classification unit B31 can usually be realized from an MPU, a memory, or the like.
- the processing procedure of the processing unit A3 or the like is usually realized by software, and the software is recorded on a recording medium such as ROM. However, it may be realized by hardware (dedicated circuit).
- Step S301 The learning unit A31 substitutes 1 for the counter i.
- Step S302 The learning unit A31 determines whether or not the i-th learning source information exists in the learning source information storage unit A11. If the i-th learning source information exists, the process goes to step S303, and if the i-th learning source information does not exist, the process goes to step S312.
- Step S303 The learning unit A31 acquires the i-th learning source information from the learning source information storage unit A11.
- Step S304 The learning unit A31 substitutes 1 for the counter j.
- Step S305 The learning unit A31 determines whether or not the j-th character exists in the character string of the i-th learning source information acquired in step S303. If the j-th character exists, the process goes to step S306, and if the j-th character does not exist, the process goes to step S309.
- Step S306 The learning unit A31 acquires the j-th character in the character string of the i-th learning source information acquired in step S303.
- Step S307 The learning unit A31 gives the j-th character acquired in step S306 to the learning module of the storage unit A1.
- the process of giving a character to the learning module is, for example, giving the character as an argument of the learning module which is a function or method, giving the character to the learning module which is an execution module, and the like, and the learning module gives the character. Is to enable learning processing using, and is widely interpreted.
- Step S309 The learning unit A31 acquires the label of the i-th learning source information acquired in step S303.
- Step S310 The learning unit A31 gives the label acquired in step S309 to the learning module of the storage unit A1.
- the process of giving the label to the learning module is, for example, giving the label as an argument of the learning module which is a function or method, giving the label to the learning module which is the execution module, and the like, and the learning module gives the label. Is to enable learning processing using, and is widely interpreted.
- Step S311 The learning unit A31 increments the counter i by 1. Return to step S302.
- Step S312 The learning unit A31 executes a learning module given a character stream and a label, and acquires a learning device.
- Step S402 The classification unit B31 substitutes 1 for the counter i.
- Step S403 The classification unit B31 determines whether or not the i-th character exists in the character string received in step S401. If the i-th character exists, the process goes to step S404, and if the i-th character does not exist, the process goes to step S407.
- Step S404 The classification unit B31 acquires the i-th character in the character string received in step S401.
- Step S405 The classification unit B31 gives the i-th character acquired in step S404 to the classification module of the storage unit B1.
- the process of giving a character to the classification module is, for example, giving the character as an argument of the classification module which is a function or a method, giving the character to the classification module which is an execution module, and the like, and the classification module gives the character. Is to enable classification processing using, and is widely interpreted.
- Step S406 The classification unit B31 increments the counter i by 1. Return to step S403.
- Step S408 The classification unit B31 executes the classification module and acquires a label.
- Step S409 The label output unit B41 outputs the label acquired in step S408. Return to step S401.
- the processing of the learning device A is the learning phase 501.
- the processing of the classification device B is the classification phase 502.
- the learning source information storage unit A11 of the learning device A now stores a large number of learning source information having the structure shown in FIG.
- the learning source information has a character string and a label.
- the label is the first label (value is "1") indicating that the number of occurrences in the point dictionary is 1, and the second label (value is "0") indicating that the number of appearances in the point dictionary is 0. ”),
- One of the third labels (value is“ 2 ”) indicating that the number of appearances in the point dictionary is 2 or more.
- map representation data processing device that acquires the direction information of the map representation data will be described.
- the area specific information acquisition unit C37 includes, for example, a first area specific information acquisition unit C371, an outer character string determination unit C372, a size information acquisition unit C373, a distance information acquisition unit C374, and a second area specific information acquisition unit C375.
- the output unit C4 includes a corresponding output unit C41, an accurate map information output unit C42, and a direction information output unit C43.
- the various types of information are, for example, a point dictionary described later, map expression data described later, accurate map information described later, a learning device, correspondence information described later, and a classification module.
- the storage unit C1 does not have to include the point dictionary storage unit C11.
- the point dictionary storage unit C11 may exist in an external device (not shown).
- the point dictionary is stored in the point dictionary storage unit C11.
- the point dictionary has one or more point position information.
- the point position information is information that associates the point information that specifies the point with the position information that specifies the position of the point.
- the point position information is, for example, information having a point information and a position information.
- the point position information is, for example, information having a pointer to the point information and a pointer to the position information.
- the data structure of the point position information does not matter.
- the map representation data is data representing a map.
- Map representation data is information in which the geographical area to be represented is limited.
- the map representation data is not accurate map information described later.
- the map representation data is, for example, data that lacks accuracy as a map.
- the map representation data is, for example, an old map, an illustration map, a rough map, a handwritten map, or the like, but the type is not limited.
- It is preferable that the map representation data is associated with a map identifier that identifies the map representation data.
- the map identifier is, for example, an ID, a file name having map representation data, a name of map representation data, or the like.
- Accurate map information is stored in the accurate map information storage unit C13.
- Accurate map information is accurate electronic map information.
- the type of map indicated by accurate map information and the data format of map information do not matter.
- the types of maps include, for example, topographic maps, geological maps, geological maps, land use maps, residential maps, route maps, road maps, guide maps, aerial photographs, satellite photographs, and the like.
- the data format of the accurate map information is, for example, raster data, vector data, KIWI format, or the like.
- the learning device of the learning device storage unit A12 is the learning device acquired by the learning device A.
- the learner has a character string having two or more characters, a label relating to the number of occurrences in the point dictionary, and one of two or three or more types of labels. It is the information acquired by giving two or more learning source information to the learning module of machine learning and executing the learning module.
- the learning device is not the learning device acquired by the learning device A, but is obtained by learning one or more positive examples that are place name information and one or more negative examples that are not place name information by a machine learning technique. It may be a learning device.
- Correspondence information storage unit C14 stores one or more correspondence information.
- the correspondence information is information indicating the correspondence between the coordinate information and the position information.
- the correspondence information may be a set of information of coordinate information and position information. Further, the correspondence information may be a set of information of a pointer to coordinate information and a pointer to position information.
- the correspondence information may correspond to the place name information, or may have the place name information.
- the data structure of the correspondence information does not matter. It is preferable that the correspondence information corresponds to the map representation data identifier that identifies the map representation data.
- the Reception department C2 receives various information and instructions.
- the various information and instructions are, for example, map representation data, operation start instructions, and output instructions.
- the operation start instruction is an instruction to start the operation of acquiring the corresponding information of the map representation data processing device C.
- the output instruction is an instruction to output map expression data or the like.
- the output instruction has, for example, a map representation data identifier that identifies the map representation data.
- the acceptance means acceptance of user input, reception from an external device (for example, a terminal device (not shown)), or the like.
- an external device for example, a terminal device (not shown)
- the input means for various information and instructions may be any, for example, a touch panel, a keyboard, a mouse, or a menu screen.
- the map representation data reception unit C21 receives the map representation data.
- acceptance means accepting information input from an input device such as a keyboard, mouse, or touch panel, receiving information transmitted via a wired or wireless communication line, and recording an optical disk, magnetic disk, semiconductor memory, or the like. It is a concept including acceptance of information read from a medium. That is, the map representation data receiving unit C21 may read the map representation data from the map representation data storage unit C12.
- the classification unit C32 determines whether or not each of one or more character strings acquired by the character string acquisition unit C31 is a character string that specifies a point.
- the classification unit C32 determines, for example, whether or not each of one or more character strings acquired by the character string acquisition unit C31 is a character string that specifies a point to be registered.
- the character string that specifies the point to be registered is, for example, point information or unique point information.
- the classification unit C32 uses, for example, one or two or more character strings acquired by the character string acquisition unit C31 and a learner, and uses machine learning techniques to obtain point information or non-point information for each of the two or more character strings. Performs classification processing to determine any of the point information.
- the classification unit C32 performs learning processing by a machine learning technique and acquires, for example, using a character string that is one or more point information as a positive example and a character string that is one or more non-point information as a negative example. Using the learned learner, a process of determining whether each of one or more character strings acquired by the character string acquisition unit C31 is point information or non-point information is performed.
- the machine learning technique may be deep learning, SVM, decision tree, random forest, etc.
- the learning module that performs the learning process and constitutes the learning device is, for example, a function in a machine learning framework such as fastText, tinySVM, TensorFlow, or various random forest functions.
- the classification module that performs the processing determined by the classification unit C32 is, for example, a function in a machine learning framework such as fastText, tinySVM, TensorFlow, or various random forest functions.
- the classification unit C32 acquires the label corresponding to each of one or more character strings acquired by the character string acquisition unit C31 by the same processing as the classification unit B31.
- the classification unit C32 gives each of two or more characters constituting the character string acquired by the character string acquisition unit C31 to the machine learning classification module in order for each character, and provides the learner of the learner storage unit A12. It is given to the classification module, the classification module is executed, and a label is acquired.
- the classification unit C32 determines, for example, that only the character string corresponding to the first label is the character string that specifies the point. Further, the classification unit C32 may determine, for example, that the character string corresponding to the first label and the third label is a character string for specifying a point.
- the grouping unit C33 determines two or more position information indicating a position closer to satisfy a predetermined condition among the three or more position information acquired by the position information acquisition unit C36. That is, the grouping unit C33 performs processing for removing the position information of points that may not be located in the map representation data.
- the position information of one or more points that will not be located in the map representation data is information indicating a position far enough to satisfy a predetermined condition with respect to other position information acquired by the position information acquisition unit C36. is there.
- the process of the grouping unit C33 is, for example, one of the following two processes. (1) Processing using representative position information
- the grouping unit C33 determines, for example, representative position information which is position information representing three or more position information acquired by the position information acquisition unit C36.
- the representative position information is usually position information indicating the central portion of the three or more position information acquired by the position information acquisition unit C36.
- the grouping unit C33 calculates, for example, the distance between each position information other than the representative position information and the representative position information, and the distance is close enough to satisfy a predetermined condition, and is representative. Acquire location information other than location information.
- the one or more position information and the representative position information are the position information determined by the grouping unit C33.
- the position information not acquired here is the excluded position information.
- Examples of the representative position information acquisition method are the following (1-1) or (1-2) or (1-3) or (1-4).
- (1-1) For example, the grouping unit C33 calculates the distance of three or more combinations of two position information out of the three or more position information acquired by the position information acquisition unit C36, and the grouping unit C33 calculates the distance of three or more combinations of the three or more combinations. Acquires two representative position information which is the shortest set which is a set of two position information corresponding to the shortest distance in the distance. (1-2) The grouping unit C33 acquires either one of the shortest sets acquired in (1-1) above as representative position information.
- the grouping unit C33 may acquire the center of gravity point of the region including the three or more position information acquired by the position information acquisition unit C36 as the representative position information.
- the grouping unit C33 may acquire the position information of the position closest to the center point in the area including the three or more position information acquired by the position information acquisition unit C36 as the representative position information. ..
- the representative position information acquisition algorithm is not limited to the above (1-1) to (1-4).
- the grouping unit C33 performs a clustering process of position information using, for example, three or more position information acquired by the position information acquisition unit C36. Then, the grouping unit C33 acquires two or more position information acquired by the clustering process. Since the clustering process is a known technique, detailed description thereof will be omitted.
- the feature location detection unit C34 detects the feature location where the feature symbol, which is a predetermined characteristic symbol, is displayed from the map representation data.
- the feature location detection unit C34 detects, for example, a feature location where a feature symbol, which is a predetermined characteristic symbol, is displayed from map representation data by an image recognition technique.
- the feature symbol is one or more kinds of symbols of an intersection or a bridge.
- a feature symbol such as an intersection or a bridge is stored in the storage unit C1, and the feature location detection unit C34, for example, shifts the pixel of the start point in the map representation data to obtain an area matching the size of the feature symbol. It is cut out from the map representation data, the similarity between the cut out area and the feature symbol of the storage unit C1 is calculated, and the area where the similarity is equal to or more than the threshold value is detected as the feature location.
- the feature location detection unit C34 performs object recognition processing of the feature symbol on the map representation data, for example, and detects the feature location. Since the object recognition process is a known technique, detailed description thereof will be omitted.
- the coordinate information acquisition unit C35 is information for specifying the coordinate position corresponding to the character string acquired by the character string acquisition unit C31, and is information for specifying the relative coordinate position in the map representation data (normally). , (X, y)).
- the coordinate position corresponding to the character string is the coordinate of a point representing the area of the character string in the map representation data, and is a representative point of the character string.
- the representative points are the center point of the character string area in the map representation data, the upper left point of the character string area in the map representation data, the lower right point of the character string area in the map representation data, and the like. ..
- the coordinate information acquisition unit C35 acquires the coordinate information that specifies the coordinate position corresponding to the character string corresponding to the position information determined by the grouping unit C33 among the character strings acquired by the character string acquisition unit C31. good. Even in such a case, it can be said that the coordinate information acquisition unit C35 acquires the coordinate information that specifies the coordinate position corresponding to the character string acquired by the character string acquisition unit C31.
- the coordinate information acquisition unit C35 acquires the coordinate information corresponding to the feature symbol.
- the coordinate information acquisition unit C35 acquires the coordinate information of the representative point of the area of the feature symbol detected by the feature location detection unit C34.
- the coordinate information acquisition unit C35 acquires, for example, the coordinate position of the center of gravity of the region in the region of the feature symbol detected by the feature location detection unit C34 as the feature location. Further, the coordinate information acquisition unit C35 acquires, for example, the coordinate position of the upper left point of the feature symbol area in the map representation data as the feature location. Further, the coordinate information acquisition unit C35 acquires, for example, the coordinate position of the lower right point of the feature symbol area in the map representation data as the feature location.
- the position information acquisition unit C36 acquires the position information corresponding to the point information which is the character string acquired by the character string acquisition unit C31 by using the point dictionary.
- the position information acquisition unit C36 acquires the position information paired with the point information which is the character string acquired by the character string acquisition unit C31 from the point dictionary. It should be noted that acquiring the position information using the point dictionary is usually to acquire the position information from the point dictionary.
- the position information acquisition unit C36 acquires, for example, the position information corresponding to the point information, which is a character string corresponding to the feature symbol, from the point dictionary.
- the character string corresponding to the feature symbol is, for example, a character string recently arranged in the feature symbol.
- the character string corresponding to the feature symbol is, for example, a character string (for example, "intersection") that is a combination of the recently arranged character string (for example, "ABC”) of the feature symbol and the name of the feature symbol (for example, "intersection”). ABC intersection ").
- the position information acquisition unit C36 is a character string recently arranged in the feature symbol, acquires a character string (for example, "ABC") on the map representation data, and names the feature symbol (for example, "intersection”). ") Is read from the storage unit C1, the two character strings are combined, a character string (for example,” ABC intersection ”) is acquired, and the position information paired with the character string (for example,” ABC intersection ”) is stored in a point dictionary. Obtained from the storage unit C11.
- the position information acquisition unit C36 may acquire the position information from the point dictionary storage unit C11 existing in an external device (not shown).
- the area identification information acquisition unit C37 is an area including a position indicated by two or more position information acquired by the position information acquisition unit C36, and may acquire area identification information for specifying an area in accurate map information.
- the second area specific information acquisition means C375 uses the distance information acquired by the distance information acquisition means C374 to specify a second area which is an area obtained by expanding the first area specified by the first area specific information. Get specific information.
- the second area specific information acquisition means C375 acquires the second area specific information that specifies the second area, which is an area expanded in each direction of the first area specific information by, for example, the distance information of each of the four directions. ..
- the second area identification information is, for example, position information of two points that specify a rectangle.
- the area specified by the first area identification information does not have to be a rectangle, and may be a polygon, a circle, or the like.
- the first angle is “0 degrees” (two points are due north and south), and the second angle is “90 degrees” (the y-axis values of the coordinate information of the two points are the same). If, the direction information acquisition unit C38 acquires the direction information "90 degrees”.
- the corresponding output unit C41 outputs the coordinate information corresponding to the character string classified into the point information by the classification unit C32 in association with the position information.
- the accurate map information output unit C42 outputs accurate map information in a manner in which the area specified by the area identification information can be known, for example.
- the output means display on a display, projection using a projector, storage on a recording medium, printing by a printer, sound output, transmission to an external device, transmission to another processing device, another program, or the like. It may be considered that the concept includes the delivery of the processing result.
- the mode in which the area specified by the area specific information can be known is to enclose the area specified by the area specific information in a frame, to change the background color of the area specified by the area specific information, and the like. Does not matter.
- the accurate map information output unit C42 outputs accurate map information in a manner in which, for example, the area specified by the two-area specific information can be known.
- the direction information output unit C43 outputs the direction information acquired by the direction information acquisition unit C38.
- the output mode of the direction information does not matter.
- the output means display on a display, projection using a projector, storage on a recording medium, printing by a printer, sound output, transmission to an external device, transmission to another processing device, another program, or the like. It may be considered that the concept includes the delivery of the processing result.
- the storage unit C1, the point dictionary storage unit C11, the map representation data storage unit C12, the accurate map information storage unit C13, the learning device storage unit A12, and the corresponding information storage unit C14 are preferably non-volatile recording media, but are volatile. It can also be realized with a sex recording medium.
- the process of storing information in the storage unit C1 or the like does not matter.
- the information may be stored in the storage unit C1 or the like via the recording medium, or the information transmitted via the communication line or the like may be stored in the storage unit C1 or the like.
- the information input via the input device may be stored in the storage unit C1 or the like.
- the reception unit C2 and the map expression data reception unit C21 can be realized by a device driver of an input means such as a touch panel or a keyboard, a menu screen control software, or the like.
- the first area specific information acquisition means C371, the outer character string determination means C372, the size information acquisition means C373, the distance information acquisition means C374, and the second area specific information acquisition means C375 can usually be realized from an MPU, a memory, or the like.
- the processing procedure of the processing unit C3 or the like is usually realized by software, and the software is recorded on a recording medium such as ROM. However, it may be realized by hardware (dedicated circuit).
- the output unit C4, the corresponding output unit C41, the accurate map information output unit C42, and the direction information output unit C43 may or may not be considered to include output devices such as displays and speakers.
- the output unit C4 can be realized by the driver software of the output device, the driver software of the output device, the output device, or the like.
- Step S801 The reception unit C2 determines whether or not the operation start instruction has been received. If the operation start instruction is accepted, the process proceeds to step S802, and if the operation start instruction is not accepted, the process proceeds to step S821.
- Step S802 The processing unit C3 substitutes 1 for the counter i.
- Step S803 The processing unit C3 determines whether or not the i-th map representation data exists in the map representation data storage unit C12. If the i-th map representation data exists, the process goes to step S804, and if it does not exist, the process returns to step S801.
- Step S804 The map expression data reception unit C21 acquires the i-th map expression data from the map expression data storage unit C12.
- Step S805 The character string acquisition unit C31 acquires one or more character strings displayed on the i-th map expression data acquired in step S804.
- the character string acquisition unit C31 acquires one or more character strings displayed on the i-th map expression data by, for example, character recognition processing.
- Step S808 The classification unit C32 performs classification processing on the j-th character string and acquires a label.
- the classification process is, for example, the process of the above-mentioned classification device B, for example, the processes of steps S402 to S409 of FIG.
- Step S809 The coordinate information acquisition unit C35 determines whether or not the j-th character string is the point information to be registered. If it is the point information to be registered, it goes to step S810, and if it is not the point information to be registered, it goes to step S812.
- the coordinate information acquisition unit C35 is an area in which the j-th character string is arranged, and acquires the coordinate information of the representative point of the area above the i-th map representation data.
- the representative point may be the center of gravity point of the region, the upper left point, the lower right point, or the like. Further, it is assumed that the information of the area where the character string is arranged can be acquired when the character string is acquired from the map representation data.
- Step S811 The position information acquisition unit C36 acquires the position information paired with the point information which is the j-th character string from the point dictionary storage unit C11. Then, the corresponding output unit C41 pairs the position information with the coordinate information acquired in step S810 and at least temporarily stores the position information in a buffer (not shown).
- Step S812 The classification unit C32 increments the counter j by 1. Return to step S807.
- Step S815) The corresponding output unit C41 determines whether or not the k-th position information exists in the position information acquired in step S813. If the k-th position information exists, the process goes to step S816, and if the k-th position information does not exist, the process goes to step S818.
- Step S816 The corresponding output unit C41 acquires the coordinate information paired with the k-th position information from a buffer (not shown). Then, the corresponding output unit C41 stores the k-th position information and the acquired coordinate information in the corresponding information storage unit C14 in association with the i-th map representation data.
- the area identification information acquisition unit C37 performs the area identification process.
- the area identification process is an area corresponding to the area of the i-th map representation data, and is a process for specifying an area in accurate map information. An example of the area specifying process will be described with reference to the flowchart of FIG.
- the direction information acquisition unit C38 performs the direction acquisition process.
- the direction acquisition process is a process of acquiring direction information for specifying the direction of the i-th map representation data. An example of the direction acquisition process will be described with reference to the flowchart of FIG.
- Step S820 The processing unit C3 increments the counter i by 1. Return to step S803.
- Step S821 The reception unit C2 determines whether or not the output instruction has been received. If the output instruction is accepted, the process proceeds to step S822, and if the output instruction is not accepted, the process returns to step S801.
- step S813 an example of the grouping process in step S813 will be described with reference to the flowchart of FIG.
- Step S901 The grouping unit C33 substitutes 1 for the counter i.
- Step S902 The grouping unit C33 determines whether or not there is a pair of the i-th two position information from the two or more position information acquired in step S811. If the i-th pair of position information exists, the process goes to step S903, and if it does not exist, the process goes to step S905.
- Step S903 The grouping unit C33 calculates the distance between the two points specified by the i-th two position information.
- Step S904 The grouping unit C33 increments the counter i by 1. Return to step S807.
- Step S905 The grouping unit C33 acquires two position information corresponding to the minimum distance among the distances between the two points calculated in step S903. Then, the grouping unit C33 acquires any one of the two position information as the representative position information.
- Step S906 The grouping unit C33 substitutes 1 for the counter j.
- Step S909 The grouping unit C33 determines whether or not the distance calculated in step S907 is close enough to satisfy a predetermined condition (for example, whether or not the distance is less than the threshold value or is less than or equal to the threshold value). ) Is judged. If the distance is short enough to satisfy the predetermined condition, the distance goes to step S910, and if the distance is not close enough to satisfy the predetermined condition, the distance goes to step S911.
- a predetermined condition for example, whether or not the distance is less than the threshold value or is less than or equal to the threshold value.
- Step S911 The grouping unit C33 increments the counter j by 1. Return to step S807.
- step S818 an example of the area identification process in step S818 will be described with reference to the flowchart of FIG.
- Step S1002 The one-area identification information acquisition means C371 acquires the position information having the maximum longitude among the position information temporarily stored in the buffer (not shown) by the grouping unit C33 in step S910.
- Step S1004 The first area identification information acquisition means C371 acquires the position information having the maximum latitude among the position information temporarily stored in the buffer (not shown) by the grouping unit C33 in step S910.
- Step S1010 The second area identification information acquisition means C375 stores the second area identification information acquired in step S1009 in association with the i-th map representation data. Return to higher processing.
- Step S1102 The direction information acquisition unit C38 acquires the direction information by using the four position information constituting the second area specific information.
- Step S1103 The direction information acquisition unit C38 stores the direction information acquired in step S1102 in association with the i-th map representation data. Return to higher processing.
- Step S1202 The output unit C4 acquires the map expression data identified by the map expression data identifier acquired in step S1201 from the map expression data storage unit C12.
- Step S1206 The output unit C4 acquires the second region identification information paired with the map representation data acquired in step S1202.
- the correspondence information storage unit C14 stores the correspondence information management table having the structure shown in FIG.
- the correspondence information management table stores one or more records having "ID”, “map representation data identifier", “coordinate information”, "position information”, and "point information”.
- the "ID” is information that identifies a record.
- the "map representation data identifier” is information that identifies the map representation data.
- Coordinat information is a coordinate value indicating a relative position on the map representation data.
- “Position information” is (latitude, longitude) here.
- Point information is a character string that has no meaning here, but is usually a point name.
- the reception unit C2 of the map representation data processing device C receives the output instruction.
- the output unit C4 performs the output processing described above.
- the map representation data of 1301 in FIG. 13 and the accurate map information of 1302 are output.
- 13021 in FIG. 13 is an area specified by the area identification information.
- 13022 is an area specified by the second area identification information.
- 13023 is a symbol for specifying the position information corresponding to the first label.
- reference numeral 13024 is a symbol that identifies the position information corresponding to the third label or the position information removed by the grouping process.
- the coordinate information such as the place name on the map representation data and the position information can be automatically acquired in association with each other.
- a character string such as an appropriate place name on the map representation data can be extracted with high accuracy, and the coordinate information such as the place name and the position information can be associated and automatically acquired.
- the position information acquisition unit that acquires the position information corresponding to the point information that is the character string acquired by the character string acquisition unit using the unit, the coordinate information acquired by the coordinate information acquisition unit, and the above.
- This is a program for functioning as a corresponding output unit that outputs the position information acquired by the position information acquisition unit in association with each other.
- FIG. 17 is a block diagram of the map representation data processing device D according to the present embodiment.
- the map representation data processing device D includes a storage unit D1, a reception unit D2, a processing unit D3, and an output unit D4.
- the storage unit D1 includes a map representation data storage unit C12 and a corresponding information storage unit C14.
- the processing unit D3 includes a current position information acquisition unit D31, a coordinate information acquisition unit D32, and a data configuration unit D33.
- the output unit D4 includes a map representation data output unit D41.
- Various information is stored in the storage unit D1.
- the various types of information are, for example, map representation data and corresponding information.
- One or two or more map expression data are stored in the map expression data storage unit C12.
- Correspondence information storage unit C14 stores two or more correspondence information. It is preferable that the correspondence information corresponds to the map representation data.
- Reception department D2 receives various information and instructions.
- the various information and instructions are, for example, operation start instructions.
- the acceptance means acceptance of user input, reception from an external device (for example, a terminal device (not shown)), or the like.
- an external device for example, a terminal device (not shown)
- the input means for various information and instructions may be any, for example, a touch panel, a keyboard, a mouse, or a menu screen.
- the processing unit D3 performs various processes.
- the various processes are, for example, processes performed by the current position information acquisition unit D31, the coordinate information acquisition unit D32, and the data configuration unit D33.
- the current position information acquisition unit D31 acquires the current position information that specifies the current position.
- the current position information acquisition unit D31 can be realized by, for example, a GPS receiver. Since the current position information acquisition unit D31 is a known technique, detailed description thereof will be omitted.
- the coordinate information acquisition unit D32 acquires the coordinate information corresponding to the current position information by using two or more corresponding information pairs with the target map representation data.
- the coordinate information is a coordinate value that specifies a relative position on the displayed map representation data.
- the coordinate information acquisition unit D32 has, for example, correspondence information 1 (coordinate information 1 (x 1 , y 1 ), position information 1 (X 1, Y 1)) and correspondence information 2 (coordinate information 2) paired with the target map representation data. (x 2, y 2), reads the position information 2 (X2, Y2)) from the correspondence information storage unit C14. Then, the coordinate information acquisition unit D32 acquires the current position information (X3, Y3) acquired by the current position information acquisition unit D31. Next, the coordinate information acquisition unit D32 calculates (X1-X2) and (Y1-Y2) from the position information 1 (X1, Y1) and the position information 2 (X2, Y2).
- the coordinate information obtaining unit D32 is, for example, (X1-X3) and calculates the number of pixels with respect to a unit distance in the x-axis, and adds the number of the pixel and the x 1, x-coordinate for the current position information (x 3 ) To get.
- the coordinate information obtaining unit D32 is, for example, (Y1-Y3) and calculates the number of pixels with respect to a unit distance in the y-axis, and adds the number of the pixels and y 1 and, y coordinate for the current position information (y 3) To get.
- the data configuration unit D33 configures the current location addition map representation data which is the map representation data in which the position indicated by the coordinate information acquired by the coordinate information acquisition unit D32 is clearly indicated on the map representation data. It should be noted that the mode of clearly indicating the position indicated by the coordinate information does not matter.
- the data configuration unit D33 arranges, for example, a predetermined symbol at a position indicated by the coordinate information.
- the map representation data output unit D41 outputs the current location additional map representation data acquired by the data configuration unit D33.
- the storage unit D1, the map representation data storage unit C12, and the corresponding information storage unit C14 are preferably non-volatile recording media, but can also be realized by volatile recording media.
- the processing unit D3, the current position information acquisition unit D31, the coordinate information acquisition unit D32, and the data configuration unit D33 can usually be realized from an MPU, a memory, or the like.
- the processing procedure of the processing unit D3 is usually realized by software, and the software is recorded on a recording medium such as ROM. However, it may be realized by hardware (dedicated circuit).
- Step S1803 The coordinate information acquisition unit D32 acquires two or more correspondence information pairs with the map representation data identifier that identifies the map representation data acquired in step S1801 from the correspondence information storage unit C14. Next, the coordinate information acquisition unit D32 acquires the coordinate information corresponding to the current position information by using the two or more corresponding information.
- Embodiment 4 In the present embodiment, three or more point names are acquired from the map representation data, coordinate information corresponding to each point name is acquired from the map representation data, and real position information corresponding to the point name is acquired from the point dictionary. Then, a map representation data processing device that outputs correspondence information, which is a set of appropriate coordinate information and position information, will be described using the coordinate information and the position information.
- an inappropriate point is determined based on the relationship information regarding the relationship between the correspondence information between the two points, and the inappropriate correspondence information corresponding to the inappropriate point is not included.
- a map representation data processing device that outputs two or more correspondence information will be described.
- distance-related information related to distance and angle-related information related to angle are used to determine an inappropriate point.
- an inappropriate point is a point corresponding to correspondence information that seems to be incorrect.
- the map representation data processing device E includes a storage unit C1, a reception unit C2, a processing unit E3, and an output unit E4.
- the relationship information acquisition unit E31 obtains relative relationship information between the corresponding information corresponding to each of the two character strings of the three or more character strings acquired by the character string acquisition unit C31 for each of the two character string pairs. It is preferable to obtain the information.
- the correspondence information acquisition unit E32 When the correspondence information acquisition unit E32 satisfies both the distance relation information and the angle relation information in a predetermined relationship, it is preferable that the correspondence information acquisition unit E32 acquires only two or more correspondence information corresponding to the distance relation information and the angle relation information. is there.
- the determination means E321 constituting the correspondence information acquisition unit E32 groups three or more relational information acquired by the relational information acquisition unit E31, and determines the relational information that does not belong to the group.
- algorithm for clustering related information for example, hierarchical aggregation clustering (Hierarchical Agglomerative Clustering [HAC]) can be used.
- HAC Hierarchical Agglomerative Clustering
- the correspondence information acquisition means E322 excludes the correspondence information corresponding to the relational information not included in the group from the correspondence information corresponding to each of the three or more character strings acquired by the character string acquisition unit C31, and the remaining two or more correspondences. Get information only.
- the area specific information acquisition unit E34 acquires, for example, representative values (for example, median value and average value) of two or more relative vertical distances for each side.
- the area specific information acquisition unit E34 acquires the position information with respect to the point of the representative value of the relative vertical distance for each side, for example.
- the area identification information acquisition unit E34 acquires the latitude or longitude of each side by, for example, "distance relationship information (MPP) x representative value of relative vertical distance" for each side.
- MPP distance relationship information
- Step S2404 The relationship information acquisition unit E31 acquires the coordinate information of each of the two points in the i-th set, and calculates the relative angle of the two points from the two coordinate information.
- Step S2506 The area specific information acquisition unit E34 acquires the j-th target information to be processed.
- Step S2508 The area identification information acquisition unit E34 uses the position information of the j-th target information, the linear distance calculated in step S2507, and the representative value (for example, MPP) of the distance-related information to the i-th side.
- the position information of each of the two end points of is acquired.
- the position information of each of these two end points is used as edge information.
- the area identification information acquisition unit E34 acquires representative values (for example, median value and average value) of two or more edge information for the i-th edge.
- the representative value is representative side information.
- the representative edge information is, for example, the information of the middle line of two or more edge information.
- the representative edge information is, for example, the average value of each of the two endpoints of the two or more edge information.
- edge information 1 edge information 1 (x 11 , y 11 ) (x 12 , y 12 ) and edge information 2 (x 21 , y 21 ) (x 22 , y 22 )
- the average value is representative edge information is ((x 11 + x 21/ 2), (y 11 + y 21/2)) ((x 12 + x 22/2), (y 12 + y 22/2)).
- step S2105 an example of the direction information acquisition process in step S2105 will be described with reference to the flowchart of FIG.
- Step S2601 The direction information acquisition unit E39 substitutes 1 for the counter i.
- Step S2701 The additional point acquisition unit E35 substitutes 1 for the counter i.
- Step S2705 The additional coordinate acquisition unit E37 acquires the coordinate information corresponding to the position information acquired in step S2703.
- Step S2707 The additional point acquisition unit E35 determines whether or not to place the point information acquired in step S2706 in the map representation data. If it is arranged, it goes to step S2708, and if it is not arranged, it goes to step S2709. The additional point acquisition unit E35 determines that the additional point acquisition unit E35 is arranged in the map representation data, for example, when one or more attribute values corresponding to the point information acquired in step S2706 satisfy a predetermined condition.
- the correspondence information acquisition unit E32 constitutes correspondence information having the correspondence information acquired in step S2703 and the coordinate information acquired in step S2705.
- Step S2710 The corresponding output unit E41 accumulates the corresponding information configured in step S2709.
- Step S2711 The counter i is incremented by 1. Return to step S2702.
- Step S2801 The output unit E4 acquires the map representation data to be processed. It is preferable that the map representation data here is data in which points have been added by the above-mentioned point addition process.
- the processing unit E3 acquires the current position information.
- the processing unit E3 may have a current position information acquisition unit D31.
- Step S2803 The processing unit E3 acquires the coordinate information corresponding to the current position information acquired in step S2802 by using the corresponding information of two or more.
- the processing unit E3 may have a coordinate information acquisition unit D32.
- Step S2804 The processing unit E3 configures the current location additional map representation data in which the position indicated by the coordinate information acquired in step S2803 is clearly indicated on the map representation data.
- the processing unit E3 may have a data configuration unit D33.
- Step S2805 The processing unit E3 acquires the area identification information acquired by the above area identification process.
- direction information and scale information may be added to the current location additional map representation data.
- map representation data of FIG. 30 is now stored in the map representation data storage unit C12.
- the map representation data is appropriately referred to as an input image.
- the map representation data reception unit C21 reads the input image from the map representation data storage unit C12.
- the character string acquisition unit C31 performs character recognition processing (OCR) on the input image, and lists the character strings “Takatsu Post Office”, “Kajigaya Station”, “2015”, “Kawasaki Citizen's Plaza”, and “Toranomon Hospital Branch”. It is assumed that the "cargo terminal" has been acquired.
- OCR character recognition processing
- the classification unit C32 determines whether or not each character string acquired by the character string acquisition unit C31 is point information by using a machine learning technique (here, LSTM Neural Network). Is not the point information, and the point information flag "-1" is written.
- LSTM Neural Network a machine learning technique
- the point information flag is a flag for determining whether or not the character string is point information.
- the location information acquisition unit C36 searches the point dictionary using each character string "Takatsu Post Office”, “Kajigaya Station”, “Kawasaki Citizen's Plaza”, “Toranomon Hospital Branch”, and "Cargo Terminal” as keys, and each character. Acquire the position information (latitude, longitude) corresponding to the column. Then, for example, the information shown in FIG. 32 can be obtained.
- the relationship information acquisition unit E31 performs the inspection process using the distance relationship information as follows.
- the related information acquisition department E31 is 2 of the C1 "Takatsu Post Office", C2 "Kajigaya Station", C3 "Kawasaki Citizen's Plaza”, C4 "Toramon Hospital Branch", and C5 "Cargo Terminal”.
- Relative distance which is the difference in coordinate information between points belonging to each set of two points, "C1-C2", “C1-C3", “C2-C4", “C3-C4", "C3-C5", and "C4-C5". (Here, the number of pixels) is calculated from the coordinate information of each point.
- the relationship information acquisition unit E31 is a point belonging to each group of two points, "C1-C2", “C1-C3", “C2-C4", “C3-C4", “C3-C5", and “C4-C5".
- the absolute distance (here, meters) which is the difference in the position information between the two, is calculated from the position information of each point.
- the relationship information acquisition unit E31 calculates the distance relationship information (here, MPP) of each set of the two points by "absolute distance / relative distance” using the relative distance and the absolute distance (FIG. 33). reference). Then, as a result of such processing, a table of distance-related information of FIG. 34 is obtained.
- the correspondence information acquisition unit E32 clusters the distance-related information shown in FIG. 34, and "0.45", “0.41”, “0.33”, “0.41”, “0.37”, and “0.44". Group and the group of “523”, “470”, “941", and “809”. Then, the corresponding information acquisition unit E32 has the point identifiers "C1” and “0.44" corresponding to the groups of "0.45", “0.41”, “0.33”, “0.41”, “0.37” and "0.44". Obtain the point identifiers "C5" corresponding to the groups "C2", “C3", “C4" and “523”, "470”, "941", and "809".
- the correspondence information acquisition unit E32 selects the point identifiers of a group in which the number of point identifiers satisfies a predetermined condition.
- the predetermined conditions are, for example, "the largest number of point identifiers", "the number of point identifiers is a threshold value (for example, 3) or more", and "the number of point identifiers is a threshold value (for example, 5)”. More. " Then, the correspondence information acquisition unit E32 acquires the point identifiers “C1”, “C2”, “C3”, and “C4”.
- the correspondence information acquisition department E32 selects "Takatsu Post Office”, “Kajigaya Station”, “Kawasaki Shimin Plaza”, and “Toranomon Hospital Branch”, and excludes (deletes) the "cargo terminal".
- a table of distance-related information of FIG. 35 is obtained. In other words, as a result of inspection processing using distance-related information, "cargo terminal" was excluded.
- the relationship information acquisition unit E31 performs an inspection process using the angle relationship information as follows.
- the related information acquisition department E31 is a set of two points "C1" out of each point of C1 "Takatsu Post Office", C2 "Kajigaya Station", C3 "Kawasaki Shimin Plaza", and C4 "Toranomon Hospital Branch”.
- -Acquire the relative angle and the absolute angle between the points belonging to each set of "C2", “C1-C3", “C2-C4", and "C3-C4".
- the relationship information acquisition unit E31 calculates the relative angle of the two points using the coordinate information of the two points forming the set.
- the relationship information acquisition unit E31 calculates the absolute angle of the two points using the position information of the two points constituting the set.
- 3601 in FIG. 36 is an image of a relative angle between C1 and C3 (an angle with reference to the base in a rectangular input image).
- 3602 in FIG. 36 is an image of an absolute angle between C1 and C3 (an angle based on the east-west direction in the real world).
- the relationship information acquisition unit E31 calculates the angle relationship information (Rotation) using the relative angle and the absolute angle.
- the related information acquisition unit E31 obtains the information shown in FIG. 37. Further, the information in which the information shown in FIG. 37 is tabulated is shown in FIG. 38. Further, the relationship information acquisition unit E31 calculates (cos (angle relationship information), sin (angle relationship information)) from the angle relationship information of FIG. 38, and calculates vector data for each set of two points ( See FIG. 39). Note that the vector data may be considered as angle-related information.
- the correspondence information acquisition unit E32 clusters the vectors of FIG. 39, and groups 1 “(cos (-2), sin (-2)) (cos (-1), sin (-2))”, and Group 2 “ (Cos (-40), sin (-40)) (cos (-24), sin (-24)) (cos (-32), sin (-32)) (cos (-26), sin (-26) )) ”Is acquired. Then, the correspondence information acquisition unit E32 obtains the point identifier "C2" corresponding to the group 1 and the point identifiers "C1", "C3", and "C4" corresponding to the group 2.
- the correspondence information acquisition unit E32 selects the point identifiers of a group in which the number of point identifiers satisfies a predetermined condition. Then, it is assumed that the correspondence information acquisition unit E32 has acquired the point identifiers "C1", “C3", and "C4". That is, the correspondence information acquisition department E32 selects "Takatsu Post Office”, “Kawasaki Shimin Plaza”, and “Toranomon Hospital Branch”, and excludes (deletes) "Kajigaya Station".
- the correspondence information acquisition unit E32 is, for example, an appropriate value of the group 2 “(cos (-40), sin (-40)) (cos (-24), sin (-24)). ) (Cos (-32), sin (-32)) (cos (-26), sin (-26)) ”, create a link between the points, and create the largest closed space point identifiers“ C1 ”and“ You may acquire "C3" and "C4". Further, the correspondence information acquisition unit E32 may, for example, create a link between points corresponding to appropriate values in group 2 and perform a process of deleting the point identifier “C2” that is not a closed space.
- the correspondence information acquisition department E32 has selected three points, "Takatsu Post Office”, “Kawasaki Shimin Plaza”, and “Toranomon Hospital Branch”, as the points of the correspondence information to be accumulated. Then, the correspondence information acquisition unit E32 acquires the correspondence information (coordinate information, position information) of each of the three points.
- the correspondence information may be linked with a character string (name of a point).
- the corresponding output unit E41 accumulates the acquired three corresponding information.
- the storage destination of the correspondence information is, for example, the correspondence information storage unit C14 or an external device (not shown).
- the correspondence information acquisition unit E32 obtains MPPs "0.41", “0.33”, and "0.44" corresponding to the three points "C1", “C3", and “C4".
- the correspondence information acquisition unit E32 calculates the representative values (here, the average value) of the three MPPs and obtains “0.39” (see FIG. 41).
- the scale acquisition unit E33 acquires the scale information (for example, “0.39”) using the representative value (MPP “0.39”) of the acquired distance-related information, and associates it with the input image. And accumulate.
- the area identification information acquisition unit E34 performs the area identification process as follows. For example, first, the area identification information acquisition unit E34 detects each side (30a, 30b, 30c, 30d) of the area of the map of the input image of FIG. 30 by the contour extraction technique.
- the technique for detecting the region by the contour extraction technique is a known technique.
- the area identification information acquisition unit E34 acquires the relative distance (here, the number of pixels) "720" from the point of the point identifier "C1" to the right side 30a.
- the relative distance is the number of pixels of the perpendicular line from the point "C1" to the right side 30a.
- the area identification information acquisition unit E34 uses the representative value (0.39) of MPP and the number of pixels "720", and performs the calculation "720 x 0.39" to obtain the absolute distance "280 m" to the right side 30a. Is calculated.
- the area identification information acquisition unit E34 uses the position information of "C1" and the absolute distance "280 m” to obtain the position information [35.5918, 139] to the right of the point "C1" in the right side 30a. .6053] has been obtained (see FIG. 43).
- the area identification information acquisition unit E34 acquires the number of pixels "1199” from the point of the point identifier "C3" to the right side 30a. Further, the area identification information acquisition unit E34 multiplies the number of pixels "1199” and the representative value of MPP (0.39) to calculate the absolute distance "468m” from the point of the point identifier "C3" to the right side 30a. To do. Further, the area identification information acquisition unit E34 uses the position information of "C3" and the absolute distance "468m” to obtain the position information [35.5849,139.] Right right of the point of "C3" in the right side 30a. 6096] is acquired (see FIG. 43). Similarly, the area identification information acquisition unit E34 acquires the position information immediately to the right of the point “C4” in the right side 30a.
- the area identification information acquisition unit E34 uses the position information and the direction information on the right side corresponding to each point of "C1", “C3", and “C4" to each place of "C1", “C3", and “C4".
- the information on the right side from the point (straight line information) [44a, 44b, 44c] is acquired.
- the conceptual diagram is shown in FIG. 44.
- the area identification information acquisition unit E34 acquires the representative value of the right side (here, the median value 44c) from the information of the three right sides.
- the representative value of this right side is determined as the boundary line of the right side in the accurate map information.
- the representative value on the right side may be an average value.
- the area specific information acquisition unit E34 performs the same processing as the acquisition process of the boundary line on the right side in the accurate map information, and performs the same processing as the acquisition process of the lower side (45a), the left side (45b), and the upper side (45c) in the accurate map information. Get border information.
- the area identification information acquisition unit E34 receives the intersections (X 1 , Y 1 ) (X 2 , Y 2 ) (X 3 , Y 3 ) (X 4 , 4 ) of the four sides (44c, 45a, 45b, 45c). Y 4 ) is obtained (see FIG. 45). The intersection is (latitude, longitude).
- the area identification information acquisition unit E34 obtains the area identification information for specifying the area of accurate map information corresponding to the area of the map in the input image.
- the area identification information includes, for example, information on four intersections, information on two points (X 1 , Y 1 ) (X 3 , Y 3 ), or information on two points (X 2 , Y 2 ) (X 4 , Y). It is the information of 4 ).
- the processing unit E3 reads out accurate map information including the area specified by the area identification information.
- the processing unit E3 configures accurate map information in which the area specified by the area identification information is specified in the accurate map information.
- the output unit E4 outputs accurate map information in which the area specified by the area identification information is specified.
- An output example of such accurate map information is shown in FIG. The area is rectangular, which is 4601.
- the additional point acquisition unit E35 searches the point dictionary and acquires one or more point position information within the area specified by the area identification information.
- the additional point acquisition unit E35 acquires the point information possessed by one or more point position information.
- the additional coordinate acquisition unit E37 acquires the coordinate information corresponding to the position information possessed by one or more point position information.
- the point addition unit E38 arranges the character string of the acquired point information on the input image at the position indicated by the acquired coordinate information for each point position information.
- the output unit E4 outputs the configured accurate map information. As described above, according to the present embodiment, it is possible to automatically acquire appropriate correspondence information in which coordinate information such as a place name on map representation data and position information are associated with each other.
- the range in the real world supported by the map representation data can be acquired.
- new point information can be added to the map representation data.
- the direction information of the map representation data can be acquired.
- map representation data processing device D described in the third embodiment can clearly indicate the current position on the map representation data by using one or two or more correspondence information automatically acquired in the present embodiment.
- the current location can be indicated on the map representation data by using two or more correspondence information automatically acquired by the map representation data processing device E.
- the processing in the present embodiment may be realized by software. Then, this software may be distributed by software download or the like. Further, this software may be recorded on a recording medium such as a CD-ROM and disseminated. It should be noted that this also applies to other embodiments herein.
- the software that realizes the information processing device in this embodiment is the following program. That is, this program uses a computer as a map representation data reception unit that accepts map representation data that is one or more types of maps of illustration maps, handwritten maps, and schematic maps, and three or more of the map representation data.
- a character string acquisition unit that acquires a character string, and a coordinate information acquisition unit that acquires coordinate information in the map representation data, which is coordinate information corresponding to each of three or more character strings acquired by the character string acquisition unit.
- Each of the three or more character strings acquired by the character string acquisition unit using a point dictionary having one or more point position information that associates the point information that specifies the point with the position information that specifies the position of the point.
- the position information acquisition unit that acquires the position information corresponding to the point information and the correspondence information that is a set of the coordinate information and the position information corresponding to each of the three or more character strings acquired by the character string acquisition unit.
- relationship information acquisition unit that acquires relationship information related to relationships and the relationship information, only two or more correspondence information corresponding to the relationship information that satisfies a predetermined relationship is acquired from the three or more correspondence information. It is a program for functioning as a correspondence information acquisition unit and a correspondence output unit for accumulating two or more correspondence information acquired by the correspondence information acquisition unit.
- FIG. 47 shows the appearance of a computer that executes the program described in the present specification to realize the map representation data processing device C and the like of the various embodiments described above.
- the above embodiments can be realized in computer hardware and computer programs running on it.
- FIG. 47 is an overview view of the computer system 300
- FIG. 48 is a block diagram of the system 300.
- the computer system 300 includes a computer 301 including a CD-ROM drive, a keyboard 302, a mouse 303, and a monitor 304.
- the computer 301 in addition to the CD-ROM drive 3012, the computer 301 is connected to the MPU 3013, the bus 3014 connected to the CD-ROM drive 3012 and the like, the ROM 3015 for storing programs such as the bootup program, and the MPU 3013. It includes a RAM 3016 that is connected and for temporarily storing instructions of an application program and providing a temporary storage space, and a hard disk 3017 for storing an application program, a system program, and data.
- the computer 301 may further include a network card that provides a connection to the LAN.
- the program for causing the computer system 300 to execute the functions of the map representation data processing device C and the like according to the above-described embodiment is stored in the CD-ROM 3101, inserted into the CD-ROM drive 3012, and further transferred to the hard disk 3017. Is also good.
- the program may be transmitted to the computer 301 via a network (not shown) and stored on the hard disk 3017.
- the program is loaded into RAM 3016 at run time.
- the program may be loaded directly from the CD-ROM3101 or network.
- the program does not necessarily include an operating system (OS) that causes the computer 301 to execute functions such as the map representation data processing device C of the above-described embodiment, or a third-party program.
- OS operating system
- the program need only include a portion of the instruction that calls the appropriate function (module) in a controlled manner to obtain the desired result. It is well known how the computer system 300 works, and detailed description thereof will be omitted.
- processing performed by hardware for example, processing performed by a modem or interface card in the transmission step (performed only by hardware). Processing is not included.
- the number of computers that execute the above program may be singular or plural. That is, centralized processing may be performed, or distributed processing may be performed.
- the two or more communication means existing in one device are physically one.
- each process may be realized by centralized processing by a single device, or may be realized by distributed processing by a plurality of devices. That is, the map representation data processing device C may operate standalone.
- the map representation data processing apparatus has an effect that coordinate information such as a place name on the map representation data and position information can be automatically acquired in association with each other, and is a map representation data processing apparatus. It is useful as such.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Library & Information Science (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Remote Sensing (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Processing Or Creating Images (AREA)
- Instructional Devices (AREA)
- Navigation (AREA)
Abstract
Description
本実施の形態において、2以上の文字を有する文字列と、文字列を分類するためのラベルとを有する2以上の学習元情報を機械学習の技法により学習し、学習器を構成する学習装置について説明する。なお、ラベルは、地点辞書での出現回数に関するラベルであり、2または2種類以上のラベルのいずれかのラベルである。また、ラベルは、出現頻度が1、0、2以上の3種類のうちのいずれかであることは好適である。なお、機械学習の技法を用いることは、機械学習のアルゴリズムを用いることと同意義である。
本実施の形態において、地図表現データから1以上の文字列を取得し、当該1以上の各文字列と対になる位置情報(緯度,経度)を取得し、当該1以上の各文字列の地図表現データ上での座標情報を取得し、当該1以上の各文字列の座標情報と位置情報とを対応付けて蓄積する地図表現データ処理装置について説明する。
(1)代表位置情報を用いた処理
(1-1)グループ化部C33は、例えば、位置情報取得部C36が取得した3以上の各位置情報のうち2つの位置情報の3以上の組み合わせの距離を算出し、当該3以上の組み合わせの距離の中の最短の距離に対応する2つの位置情報の組である最短組である2つの代表位置情報を取得する。
(1-2)グループ化部C33は、上記(1-1)で取得した最短組の中のどちらか一方を代表位置情報として取得する。
(1-3)グループ化部C33は、位置情報取得部C36が取得した3以上の位置情報を含む領域の重心点を代表位置情報として取得しても良い。
(1-4)グループ化部C33は、位置情報取得部C36が取得した3以上の位置情報が含まれる領域の中の中心点に最も近い位置の位置情報を代表位置情報として取得しても良い。
(2)クラスタリング処理
本実施の形態において、実施の形態2で取得した対応情報を用いて、地図表現データ上に現在位置を明示できる地図表現データ処理装置Dについて説明する。
本実施の形態において、地図表現データから、3以上の地点名を取得し、当該各地点名に対応する座標情報を地図表現データから取得し、地点名に対応するリアルな位置情報を地点辞書から取得し、座標情報と位置情報とを用いて、適切な座標情報と位置情報との組である対応情報を出力する地図表現データ処理装置について説明する。
(cos(-40),sin(-40))(cos(-24),sin(-24))(cos(-32),sin(-32))(cos(-26),sin(-26))」を取得する。そして、対応情報取得部E32は、グループ1に対応する地点識別子「C2」と、グループ2に対応する地点識別子「C1」「C3」「C4」を得る。
B 分類装置
C,D,E 地図表現データ処理装置
A1,B1,C1,D1 格納部
A2,B2,C2,D2 受付部
A3,B3,C3,D3,E3 処理部
A4,B4,C4,D4、E4 出力部
A11 学習元情報格納部
A12 学習器格納部
A31 学習部
A41 蓄積部
B21 文字列受付部
B31 分類部
B41 ラベル出力部
C11 地点辞書格納部
C12 地図表現データ格納部
C13 正確地図情報格納部
C14 対応情報格納部
C21 地図表現データ受付部
C31 文字列取得部
C32 分類部
C33 グループ化部
C34 特徴箇所検出部
C35 座標情報取得部
C36 位置情報取得部
C37 領域特定情報取得部
C38 方角情報取得部
C41 対応出力部
C42 正確地図情報出力部
C43 方角情報出力部
C371 第一領域特定情報取得手段
C372 外側文字列決定手段
C373 サイズ情報取得手段
C374 距離情報取得手段
C375 第二領域特定情報取得手段
D31 現在位置情報取得部
D32 座標情報取得部
D33 データ構成部
D41 地図表現データ出力部
E31 関係情報取得部
E32 対応情報取得部
E33 縮尺取得部
E34 領域特定情報取得部
E35 追加地点取得部
E36 追加位置取得部
E37 追加座標取得部
E38 地点追記部
E39 方角情報取得部
E41 対応出力部
E42 縮尺情報出力部
E43 領域特定情報出力部
E44 方角情報出力部
E321 判断手段
E322 対応情報取得手段
Claims (19)
- イラスト地図、手書き地図、略地図のうちのいずれか1以上の種類の地図である地図表現データを受け付ける地図表現データ受付部と、
前記地図表現データから文字列を取得する文字列取得部と、
前記文字列取得部が取得した文字列に対応する座標位置を特定する情報であり、前記地図表現データの中の相対的な座標位置を特定する情報である座標情報を取得する座標情報取得部と、
地点を特定する地点情報と当該地点の位置を特定する位置情報とを対応付ける情報である1以上の地点位置情報を有する地点辞書を用いて、前記文字列取得部が取得した文字列である地点情報に対応する位置情報を取得する位置情報取得部と、
前記座標情報取得部が取得した座標情報と前記位置情報取得部が取得した位置情報とを対応付けて出力する対応出力部とを具備する地図表現データ処理装置。 - 前記文字列取得部は、
前記地図表現データから2以上の文字列を取得し、
前記文字列取得部が取得した2以上の各文字列が地点を特定する文字列であるか否かを判断する分類部をさらに具備し、
前記対応出力部は、
前記分類部が地点を特定する文字列であると判断した文字列に対応する座標情報と位置情報とを対応付けて出力する請求項1記載の地図表現データ処理装置。 - 2以上の文字を有する文字列と、地点辞書での出現回数に関するラベルであり、2または3種類以上のラベルのうちのいずれかのラベルとを有する2以上の学習元情報を機械学習の学習モジュールに与え、当該学習モジュールを実行し、取得された学習器が格納される学習器格納部をさらに具備し、
前記分類部は、
前記文字列取得部が取得した2以上の各文字列と前記学習器とを用いて、機械学習の技法により前記2以上の各文字列に対して、地点情報または非地点情報のいずれかを決める分類処理を行い、
前記対応出力部は、
前記分類部が地点情報に分類した文字列に対応する座標情報と位置情報とを対応付けて出力する請求項2記載の地図表現データ処理装置。 - 前記地図表現データから、予め決められた特徴的な図柄である特徴図柄が表出している箇所である特徴箇所を検出する特徴箇所検出部をさらに具備し、
前記座標情報取得部は、
前記特徴図柄に対応する座標情報を取得し、
前記位置情報取得部は、
前記地点辞書を用いて、前記特徴図柄に対応する文字列である地点情報に対応する位置情報を取得する請求項1から請求項3いずれか一項に記載の地図表現データ処理装置。 - 前記文字列取得部は、
前記地図表現データから3以上の文字列を取得し、
前記位置情報取得部が取得した3以上の位置情報のうち、予め決められた条件を満たすほど近い位置を示す2以上の位置情報を決定するグループ化部をさらに具備し、
前記対応出力部は、
前記グループ化部が決定した2以上の各位置情報と前記座標情報取得部が取得した2以上の各座標情報とを対応付けて出力する請求項1から請求項4いずれか一項に記載の地図表現データ処理装置。 - 正確な地図情報である正確地図情報が格納される正確地図情報格納部と、
前記グループ化部が決定した2以上の位置情報が示す位置を含む領域であり、前記正確地図情報の中の領域を特定する領域特定情報を取得する領域特定情報取得部と、
前記領域特定情報が特定する領域が分かる態様で、前記正確地図情報を出力する正確地図情報出力部とをさらに具備する請求項5記載の地図表現データ処理装置。 - 前記領域特定情報取得部は、
前記グループ化部が決定した2以上の位置情報が示す位置を含む領域であり、前記正確地図情報の中の領域である第一領域を特定する第一領域特定情報を取得する第一領域特定情報取得手段と、
前記位置情報取得部が取得した位置情報のうち、最も外側に位置する1以上の各位置情報に対応する文字列を決定する外側文字列決定手段と、
前記外側文字列決定手段が決定した1以上の文字列の外側であり、前記地図表現データの隅までのサイズに関するサイズ情報を取得するサイズ情報取得手段と、
前記座標情報取得部が取得した座標情報と前記位置情報取得部が取得した位置情報との組を2組以上用いて、前記サイズ情報取得手段が取得したサイズ情報に対応する距離を特定する距離情報を取得する距離情報取得手段と、
前記距離情報を用いて、前記第一領域特定情報が特定する第一領域を広げた領域である第二領域を特定する第二領域特定情報を取得する第二領域特定情報取得手段とを具備し、
前記正確地図情報出力部は、
前記第二領域特定情報が特定する領域が分かる態様で、前記正確地図情報を出力する請求項6記載の地図表現データ処理装置。 - 前記領域特定情報または前記第二領域特定情報は、矩形を構成する2点の情報を含み、
前記領域特定情報または前記第二領域特定情報を用いて、前記地図表現データの方角に関する方角情報を取得する方角情報取得部と、
前記方角情報を出力する方角情報出力部とをさらに具備する請求項7記載の地図表現データ処理装置。 - 前記文字列取得部が取得した3以上の各文字列に対応する前記座標情報と前記位置情報との組である対応情報の間の関係に関する関係情報を取得する関係情報取得部と、
前記関係情報を用いて、前記3以上の対応情報の中で、予め決められた関係を満たす関係情報に対応する2以上の対応情報のみを取得する対応情報取得部と、
前記対応情報取得部が取得した2以上の対応情報を蓄積する対応出力部とをさらに具備する請求項1から請求項4いずれか一項に記載の地図表現データ処理装置。 - 前記関係情報取得部は、
前記文字列取得部が取得した3以上の文字列のうちの2つの各文字列に対応する対応情報の間の相対的な関係情報を、2つの文字列の組ごとに取得し、
前記対応情報取得部は、
前記関係情報取得部が取得した3以上の関係情報をグループ化し、いずれのグループにも入らない関係情報を決定する判断手段と、
前記文字列取得部が取得した3以上の各文字列に対応する対応情報の中から、前記グループに入らない関係情報に対応する対応情報を除き、残った2以上の対応情報のみを取得する対応情報取得手段とを具備する請求項9記載の地図表現データ処理装置。 - 前記関係情報取得部は、
前記2つの文字列の組ごとに、前記2つの文字列の座標情報の差異である相対距離と前記2つの文字列の位置情報の差異である絶対距離との関係を示す関係情報である距離関係情報を取得し、
前記対応情報取得部は、
前記3以上の各距離関係情報が予め決められた関係を満たすか否かを判断し、前記文字列取得部が取得した3以上の各文字列に対応する対応情報の中から、前記予め決められた関係を満たさない距離関係情報に対応する対応情報を除き、残った2以上の対応情報のみを取得する請求項9または請求項10記載の地図表現データ処理装置。 - 前記3以上の距離関係情報から距離関係情報の代表値を取得し、当該距離関係情報の代表値を用いて、地図表現データの縮尺情報を取得する縮尺取得部と、
前記縮尺情報を出力する縮尺情報出力部とをさらに具備する請求項11記載の地図表現データ処理装置。 - 前記文字列取得部が取得した文字列に対応する座標情報と前記地図表現データの4辺のうちの各辺との垂直距離である相対垂直距離を取得し、当該相対垂直距離と前記縮尺情報とを用いて、前記地図表現データの実世界における範囲を特定する領域特定情報を取得する領域特定情報取得部と、
前記領域特定情報を出力する領域特定情報出力部とをさらに具備する請求項12記載の地図表現データ処理装置。 - 前記領域特定情報が示す範囲の地点情報を前記地点辞書から取得する追加地点取得部と、
前記地点情報に対応する位置情報を前記地点辞書から取得する追加位置取得部と、
前記位置情報に対応する座標情報を取得する追加座標取得部と、
前記地図表現データの上の位置であり、前記座標情報が示す位置に前記地点情報を配置する地点追記部とをさらに具備する請求項13記載の地図表現データ処理装置。 - 前記関係情報取得部は、
前記2つの文字列の組ごとに、前記2つの文字列の座標情報から得られる角度である相対角度と前記2つの文字列の位置情報から得られる角度である絶対角度との関係を示す関係情報である角度関係情報を取得し、
前記対応情報取得部は、
前記3以上の各角度関係情報が予め決められた関係を満たすか否かを判断し、前記文字列取得部が取得した3以上の各文字列に対応する対応情報の中から、前記予め決められた関係を満たさない角度関係情報に対応する対応情報を除き、残った2以上の対応情報のみを取得する請求項9または請求項10記載の地図表現データ処理装置。 - 前記3以上の角度関係情報から角度関係情報の代表値を取得し、当該角度関係情報の代表値を用いて、地図表現データの方角を示す方角情報を取得する方角情報取得部と、
前記方角情報を出力する方角情報出力部とをさらに具備する請求項15記載の地図表現データ処理装置。 - 地図表現データが格納される地図表現データ格納部と、
請求項1から請求項16いずれか一項に記載の地図表現データ処理装置が出力した座標情報と位置情報との組である2以上の対応情報が格納される対応情報格納部と、
現在位置を特定する現在位置情報を取得する現在位置情報取得部と、
前記2以上の対応情報を用いて、前記現在位置情報に対応する座標情報を取得する座標情報取得部と、
前記地図表現データの上に、前記座標情報が示す位置が明示された地図表現データである現在地付加地図表現データを構成するデータ構成部と、
前記現在地付加地図表現データを出力する地図表現データ出力部とを具備する地図表現データ処理装置。 - 地図表現データ受付部と、文字列取得部と、座標情報取得部と、位置情報取得部と、対応出力部とにより実現される対応情報の生産方法であって、
前記地図表現データ受付部が、イラスト地図、手書き地図、略地図のうちのいずれか1以上の種類の地図である地図表現データを受け付ける地図表現データ受付ステップと、
前記文字列取得部が、前記地図表現データから文字列を取得する文字列取得ステップと、
前記座標情報取得部が、前記文字列取得ステップで取得された文字列に対応する座標位置を特定する情報であり、前記地図表現データの中の相対的な座標位置を特定する情報である座標情報を取得する座標情報取得ステップと、
前記位置情報取得部が、地点を特定する地点情報と当該地点の位置を特定する位置情報とを対応付ける情報である1以上の地点位置情報を有する地点辞書を用いて、前記文字列取得ステップで取得された文字列である地点情報に対応する位置情報を取得する位置情報取得ステップと、
前記対応出力部が、前記座標情報取得ステップで取得され座標情報と前記位置情報取得ステップで取得された位置情報とを対応付けて出力する対応出力ステップとを具備する対応情報の生産方法。 - 地点を特定する地点情報と当該地点の位置を特定する位置情報とを対応付ける情報である1以上の地点位置情報を有する地点辞書が格納される地点辞書格納部にアクセス可能なコンピュータを、
イラスト地図、手書き地図、略地図のうちのいずれか1以上の種類の地図である地図表現データを受け付ける地図表現データ受付部と、
前記地図表現データから文字列を取得する文字列取得部と、
前記文字列取得部が取得した文字列に対応する座標位置を特定する情報であり、前記地図表現データの中の相対的な座標位置を特定する情報である座標情報を取得する座標情報取得部と、
前記地点辞書を用いて、前記文字列取得部が取得した文字列である地点情報に対応する位置情報を取得する位置情報取得部と、
前記座標情報取得部が取得した座標情報と前記位置情報取得部が取得した位置情報とを対応付けて出力する対応出力部として機能させるためのプログラム。
Priority Applications (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US17/622,052 US12148179B2 (en) | 2019-06-25 | 2020-06-22 | Map representation data processing device, correspondence information production method, and program |
| JP2021526978A JP7513990B2 (ja) | 2019-06-25 | 2020-06-22 | 地図表現データ処理装置、対応情報の生産方法、およびプログラム |
| CN202080046781.1A CN114008608B (zh) | 2019-06-25 | 2020-06-22 | 地图表现数据处理装置、对应信息的产生方法、以及程序 |
| EP20831278.5A EP3992952B1 (en) | 2019-06-25 | 2020-06-22 | Map representation data processing device, correspondence information production method, and program |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2019117259 | 2019-06-25 | ||
| JP2019-117259 | 2019-06-25 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2020262312A1 true WO2020262312A1 (ja) | 2020-12-30 |
Family
ID=74060893
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/JP2020/024435 Ceased WO2020262312A1 (ja) | 2019-06-25 | 2020-06-22 | 地図表現データ処理装置、対応情報の生産方法、およびプログラム |
| PCT/JP2020/024436 Ceased WO2020262313A1 (ja) | 2019-06-25 | 2020-06-22 | 地図表現データ処理装置、情報処理方法、およびプログラム |
Family Applications After (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/JP2020/024436 Ceased WO2020262313A1 (ja) | 2019-06-25 | 2020-06-22 | 地図表現データ処理装置、情報処理方法、およびプログラム |
Country Status (5)
| Country | Link |
|---|---|
| US (2) | US12148179B2 (ja) |
| EP (2) | EP3992952B1 (ja) |
| JP (3) | JP7513990B2 (ja) |
| CN (2) | CN114008608B (ja) |
| WO (2) | WO2020262312A1 (ja) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2024069820A (ja) * | 2022-11-10 | 2024-05-22 | Lineヤフー株式会社 | アプリケーションプログラム、情報処理システム、および情報処理方法 |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117911573B (zh) * | 2024-01-25 | 2024-10-11 | 中国南方航空股份有限公司 | 一种多边形空域绘制方法、装置、设备及存储介质 |
| JP7671545B1 (ja) | 2024-07-02 | 2025-05-02 | 株式会社ビジネス・プログレス | 地図管理システム及び地図管理方法 |
| CN119669544B (zh) * | 2025-02-19 | 2025-06-27 | 国网江西省电力有限公司供电服务管理中心 | 计量装置实时定位方法、装置、电子设备及介质 |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010272082A (ja) * | 2009-05-25 | 2010-12-02 | Hitachi Ltd | 情報共有システム、情報共有方法、および情報共有プログラム |
| JP2011043788A (ja) * | 2009-07-21 | 2011-03-03 | Advanced Telecommunication Research Institute International | 情報処理装置、情報処理方法、およびプログラム |
| JP2011158518A (ja) | 2010-01-29 | 2011-08-18 | Advanced Telecommunication Research Institute International | 地図情報システム、端末装置、サーバ装置およびプログラム |
| JP2012168069A (ja) * | 2011-02-16 | 2012-09-06 | Mic Ware:Kk | 地図情報処理装置、ナビゲーション装置、地図情報処理方法、およびプログラム |
| JP2014032308A (ja) * | 2012-08-03 | 2014-02-20 | Micware Co Ltd | 地図情報処理装置、地図情報処理システム、地図情報処理方法、およびプログラム |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002169823A (ja) | 2000-12-01 | 2002-06-14 | Minolta Co Ltd | データベース検索プログラムを記録したコンピュータ読み取り可能な記録媒体、データベース検索方法およびデータベース検索装置 |
| JP5832432B2 (ja) * | 2010-06-15 | 2015-12-16 | 株式会社ナビタイムジャパン | ナビゲーションシステム、ナビゲーション方法、および、プログラム |
| JP2012185741A (ja) | 2011-03-07 | 2012-09-27 | Ricoh Co Ltd | 画像形成装置、情報処理装置、画像処理システム、及びプログラム |
| US8825392B2 (en) * | 2011-06-30 | 2014-09-02 | Navteq B.V. | Map view |
| DE112011105547B4 (de) * | 2011-08-24 | 2021-12-23 | Mitsubishi Electric Corporation | Navigationsvorrichtung |
| JP7290281B2 (ja) | 2017-10-03 | 2023-06-13 | 株式会社Stroly | 情報処理装置、情報システム、情報処理方法、およびプログラム |
| CN109033465A (zh) * | 2018-08-31 | 2018-12-18 | 北京诸葛找房信息技术有限公司 | 基于地理位置与名字相似的多平台小区合并方法 |
-
2020
- 2020-06-22 WO PCT/JP2020/024435 patent/WO2020262312A1/ja not_active Ceased
- 2020-06-22 CN CN202080046781.1A patent/CN114008608B/zh active Active
- 2020-06-22 JP JP2021526978A patent/JP7513990B2/ja active Active
- 2020-06-22 US US17/622,052 patent/US12148179B2/en active Active
- 2020-06-22 US US17/622,473 patent/US11928835B2/en active Active
- 2020-06-22 EP EP20831278.5A patent/EP3992952B1/en active Active
- 2020-06-22 WO PCT/JP2020/024436 patent/WO2020262313A1/ja not_active Ceased
- 2020-06-22 CN CN202080046773.7A patent/CN114080600B/zh active Active
- 2020-06-22 EP EP20831685.1A patent/EP3992953B1/en active Active
- 2020-06-22 JP JP2021526979A patent/JP7521760B2/ja active Active
-
2024
- 2024-07-03 JP JP2024107267A patent/JP7736345B2/ja active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010272082A (ja) * | 2009-05-25 | 2010-12-02 | Hitachi Ltd | 情報共有システム、情報共有方法、および情報共有プログラム |
| JP2011043788A (ja) * | 2009-07-21 | 2011-03-03 | Advanced Telecommunication Research Institute International | 情報処理装置、情報処理方法、およびプログラム |
| JP2011158518A (ja) | 2010-01-29 | 2011-08-18 | Advanced Telecommunication Research Institute International | 地図情報システム、端末装置、サーバ装置およびプログラム |
| JP2012168069A (ja) * | 2011-02-16 | 2012-09-06 | Mic Ware:Kk | 地図情報処理装置、ナビゲーション装置、地図情報処理方法、およびプログラム |
| JP2014032308A (ja) * | 2012-08-03 | 2014-02-20 | Micware Co Ltd | 地図情報処理装置、地図情報処理システム、地図情報処理方法、およびプログラム |
Non-Patent Citations (1)
| Title |
|---|
| See also references of EP3992952A4 |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2024069820A (ja) * | 2022-11-10 | 2024-05-22 | Lineヤフー株式会社 | アプリケーションプログラム、情報処理システム、および情報処理方法 |
| JP7739251B2 (ja) | 2022-11-10 | 2025-09-16 | Lineヤフー株式会社 | アプリケーションプログラム、情報処理システム、および情報処理方法 |
Also Published As
| Publication number | Publication date |
|---|---|
| US12148179B2 (en) | 2024-11-19 |
| CN114080600A (zh) | 2022-02-22 |
| JP2024138342A (ja) | 2024-10-08 |
| JPWO2020262312A1 (ja) | 2020-12-30 |
| WO2020262313A1 (ja) | 2020-12-30 |
| EP3992952C0 (en) | 2025-09-24 |
| US20220254053A1 (en) | 2022-08-11 |
| US20220318297A1 (en) | 2022-10-06 |
| JP7513990B2 (ja) | 2024-07-10 |
| EP3992953C0 (en) | 2024-12-04 |
| CN114008608B (zh) | 2026-04-10 |
| CN114080600B (zh) | 2026-02-17 |
| JP7736345B2 (ja) | 2025-09-09 |
| EP3992952A1 (en) | 2022-05-04 |
| EP3992953B1 (en) | 2024-12-04 |
| EP3992952B1 (en) | 2025-09-24 |
| EP3992953A4 (en) | 2022-08-03 |
| EP3992953A1 (en) | 2022-05-04 |
| JP7521760B2 (ja) | 2024-07-24 |
| JPWO2020262313A1 (ja) | 2020-12-30 |
| US11928835B2 (en) | 2024-03-12 |
| EP3992952A4 (en) | 2022-08-10 |
| CN114008608A (zh) | 2022-02-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11580755B2 (en) | Method, apparatus, and system for determining polyline homogeneity | |
| EP3992952B1 (en) | Map representation data processing device, correspondence information production method, and program | |
| CN111542860B (zh) | 用于自主车辆的高清地图的标志和车道创建 | |
| US11392797B2 (en) | Method, apparatus, and system for filtering imagery to train a feature detection model | |
| EP2273337B1 (en) | Generating a graphic model of a geographic object and systems thereof | |
| CN110378293B (zh) | 一种基于实景三维模型生产高精度地图的方法 | |
| JP5797419B2 (ja) | 地図情報処理装置、ナビゲーション装置、地図情報処理方法、およびプログラム | |
| Li et al. | Urban infrastructure audit: An effective protocol to digitize signalized intersections by mining street view images | |
| US10776951B2 (en) | Method, apparatus, and system for an asymmetric evaluation of polygon similarity | |
| Agrawala | Visualizing route maps | |
| CN116863427B (zh) | 基于胶囊网络的车辆视觉重定位方法、设备及介质 | |
| EP4273502A1 (en) | Method, apparatus, and computer program product for map geometry generation based on data aggregation and conflation | |
| CN119600608B (zh) | 一种城市建筑物的语义实景三维建模方法 | |
| CN112686988A (zh) | 三维建模方法、装置、电子设备及存储介质 | |
| EP4300043A1 (en) | Method and apparatus for providing a path-based map matcher | |
| JP2021005143A (ja) | 学習装置、分類装置、学習器の生産方法、分類方法、およびプログラム | |
| CN113157832A (zh) | 一种基于边缘计算的智能汽车高精度地图生成方法 | |
| US12292308B2 (en) | Method, apparatus, and computer program product for map geometry generation based on object detection | |
| Edelkamp | Navigation Data | |
| Cheng Sr et al. | Unite segmentation method of uneven sparse road point cloud considering multi-sampling radius | |
| Gotlib | Mobile maps-modelling of cartographic presentation | |
| JP7035289B2 (ja) | ガイド情報提案装置、およびプログラム | |
| Kitamoto et al. | Geometric correction of measured historical maps with a pixel-oriented and Geobrowser-friendly framework | |
| Aijazi | 3D urban cartography incorporating recognition and temporal integration | |
| Bartelme | Geographic In 6 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20831278 Country of ref document: EP Kind code of ref document: A1 |
|
| ENP | Entry into the national phase |
Ref document number: 2021526978 Country of ref document: JP Kind code of ref document: A |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| ENP | Entry into the national phase |
Ref document number: 2020831278 Country of ref document: EP Effective date: 20220125 |
|
| WWG | Wipo information: grant in national office |
Ref document number: 2020831278 Country of ref document: EP |