JP5253675B2 - Character string arrangement device, character string arrangement candidate decision method, and character string arrangement candidate decision program - Google Patents

Abstract

A character string arrangement device is provided with: a character string data acquisition unit (2) that acquires data for character strings to be arranged on a display screen; a unit for creating character string arrangement candidate starting points (3) that creates character string arrangement candidate starting points on the display screen where the arrangement of the character strings acquired by the character string data acquisition unit (2) are to start; a unit for calculating minimum cost function values (4) that arranges the character strings in such a manner that the reading of each character string arrangement candidate starting point created by the unit for creating character string arrangement candidate starting points (3) as a starting point proceeds in a forward direction based on the direction in which the character string arrangement candidate starting points face each other on the display screen, and that calculates cost function values for evaluating factors that affect the readability of the character strings on the basis of the positional relationships of the character strings for each arrangement; and a unit for storing output character string arrangement candidate starting points (5) for storing the candidate positions having a minimum cost function value as character string arrangement candidate starting points when changing the arrangement start position of character strings.

Description

  The present invention relates to a character string arrangement device, a character string arrangement candidate determination method, and a character string arrangement candidate determination program that arrange and display a character string such as a place name on a map screen.

  As a conventional technique for arranging and displaying a character string such as a place name on a map screen, there is one disclosed in Patent Document 1, for example. In Patent Document 1, when a character string of a name such as a road, river, or railway is arranged along a link that represents a road, river, or railway on a map screen, the character string is displayed on the screen in consideration of the readability of the character string. The character string arrangement start point at which to start the arrangement is determined.

Shawn Edmondson, “A General Cartographic Labeling Algorithm”, The International Journal for Geographic Information and Geovisualization, Volume 33, Number 4 / Winter 1996

  In the conventional technique represented by Patent Document 1, when displaying a character example such as a road name, a road is represented by connecting a plurality of road links 101 on a display screen 100 as shown in FIG. A character string 102 indicating information such as the name of the road is arranged along the road link 101 starting from a character string arrangement start point 103 provided on the road link 101. In this case, depending on the position where the character string arrangement start point 103 is provided, the character strings overlap each other or the character strings must be deformed along the road link, and the readability of the character strings is lowered.

For example, in FIG. 2, there is a road that intersects the road indicated by the road link 101 in FIG. 1, and the character string 102 a is arranged along the road link 101 of the road from the character arrangement start point 103 a as a starting point. The case where 102 and the character string 102a overlap is shown. In this case, the overlapped characters are very difficult to read and readability deteriorates.
Further, FIG. 3 represents the curved portion of the road by connecting adjacent road links 101 with a certain angle, and the character string arrangement start point 103A is provided on the road link 101 near the curved portion. , A case where the character string 102 is deformed along the road link 101 is shown. Also in this case, the continuity of the characters in the character string is impaired, the readability is deteriorated, and it is difficult to grasp the contents of the character string at a glance.

Therefore, in the conventional technique represented by Patent Document 1, the overlap between character strings and the degree of curve of the road along which the character strings follow are converted into functions, respectively, that is, the position where the characteristic indicated by these function values is the smallest, that is, the character A position where the overlap between columns and the degree of curve of the road does not occur is specified, and the character string arrangement start point is changed to this position.
FIG. 4 is a diagram for explaining a function indicating duplication of character strings. In the example shown in FIG. 4A, a plurality of roads intersect with each other, and the character strings 102, 102a, 102b are replaced by character string arrangement start points 103, 103a, 103b along the road links 101 of the respective roads. Each is arranged as a starting point.
The function overlap (i) shown in FIG. 4B is a function representing the number of overlapping character strings in FIG. For example, since the character strings 102 and 102b overlap with only the character string 102a, respectively, overlap (i) = 1, and since the character string 102a overlaps both the character strings 102 and 102b, overlap (i) = 2. .

FIG. 5 is a diagram for explaining a function that indicates the degree of curve of the road along which the character string follows. As shown in FIG. 5, by obtaining a linear approximation AveLine (i) of a road node on a road in which a plurality of road links are connected via road nodes, it is expressed by AveLine (i) indicated by a broken line in FIG. 5. 5 is a value obtained by dividing the sum of the areas indicated by the solid triangles in FIG. 5 by the length of the straight line indicated by AveLine (i). i).
Using these functions overlap (i) and flat (i), the character string arrangement start point is changed to a position where the value of the cost function f is minimized as shown in the following equation (1).
The character string arrangement start point changing process is normally performed offline, and the changed character string arrangement start point is stored in the storage device. On-line, character strings are arranged starting from the character string arrangement start point stored in the storage device. In the following formula (1), α ₁ and α ₂ are parameters for adjusting the degree of contribution of the values of the functions overlap (i) and flat (i) to the cost function f.
f = α ₁ overlap (i) + α ₂ flat (i) (1)

  Moreover, the conventional technique represented by patent document 1 presupposes the case where the arrangement | positioning of a character string does not change even if the direction of a map changes. For example, as shown in FIG. 6A, when the map is oriented north (north up) and the character string 102 is arranged along the north side of the road link 101, as shown in FIG. Even if the direction of the map is changed to south (south up), the character string 102 is similarly arranged on the north side of the road link 101.

  On the other hand, in an actual car navigation system, the direction of the map may change according to the traveling direction of the own vehicle (heading up). For example, as shown in FIG. 7 (a), if the vehicle is traveling toward the north, the direction of the map is north, and the traveling direction of the vehicle is southwest as shown in FIG. 7 (b). If it changes, the orientation of the map will be southwest. In this case, in the conventional car navigation system, in order to maintain the readability of the character string, the character is displayed upward so that it can be read to the left in any map orientation.

For example, as shown in FIG. 8A, even if one of the both sides of the road link 101 is in the direction of the map, the character string 102 arranged along the road link 101 is directed upward. The character string arrangement patterns of the character strings 102 having different directions are stored as patterns 1 and 2 corresponding to both sides of the road link 101. Here, as shown in FIG. 8 (b), if the vehicle is traveling north, the map direction is north. Therefore, by using the character string arrangement of pattern 1, the character string 102 is arranged upward. Is done. Further, if the orientation of the map is south, the character string arrangement of pattern 2 is switched as shown in FIG.
Thus, when switching the arrangement of the character string according to the direction of the map, there is a problem in that the readability of the character string may be impaired when the conventional technique represented by Patent Document 1 is applied.

Conventionally, the arrangement position of a character string is changed to a position having a high readability for a character string arrangement in a certain one-way map direction.
In this case, for example, as shown in FIG. 9 (a), the character strings 102 and 102a do not overlap if the arrangement of the character strings is changed when the map orientation is north, but as shown in FIG. 9 (b), When the direction of the map changes to the south, the arrangement of the character string 102a is switched upward so that the character can be read to the left. Therefore, the character string 102 and the character string 102a overlap.
In order to avoid such inconveniences, it is sufficient to determine the character string arrangement start point for each direction of the map so that duplication of character strings and deformation display of the character string do not occur. It is necessary to prepare as many pieces of information about the character string arrangement start point as the number of map directions, and a huge storage area is required.

  The present invention has been made to solve the above-described problems, and does not cause an increase in the amount of information to be prepared and can be arranged on a display screen without losing readability. It is an object to obtain a device, a character string arrangement candidate determination method, and a character string arrangement candidate determination program.

  A character string arrangement device according to the present invention includes a character string data acquisition unit that acquires data of a character string to be arranged on a display screen, and a character string acquisition unit that starts arrangement of the character string acquired by the character string data acquisition unit. Position candidate creation section that creates position candidates, and character strings are arranged in such a way that readings are forward in opposite directions on the display screen starting from the position candidates created by the position candidate creation section. A cost function value calculating unit that calculates a cost function value that evaluates a factor that affects the readability of the character string based on the positional relationship between the character strings, and a position candidate having the smallest cost function value And a storage unit that stores a position candidate when changing the arrangement start position.

  According to this invention, even if the direction on the display screen changes, there is an effect that the character string can be arranged without impairing the readability without causing an increase in the amount of information to be prepared.

It is a figure which shows the example of arrangement | positioning of a character string. It is a figure which shows an example in case character strings overlap. It is a figure which shows an example in case a character string is deform | transformed. It is a figure explaining the function which shows duplication of character strings. It is a figure explaining the function which shows the curve degree of the road which a character string follows. It is a figure which shows the case where the arrangement | positioning of a character string does not change even if the direction of a map changes. It is a figure which shows the map display in case the direction of a map turns into the advancing direction of the own vehicle. It is a figure which shows an example of the process which always displays a character string upwards. It is a figure explaining the arrangement | positioning change of the character string accompanying the change of the direction of a map. It is a block diagram which shows the structure of the character string arrangement | positioning apparatus by Embodiment 1 of this invention. It is a block diagram which shows the structure of a character string arrangement | positioning start candidate point production part. It is a block diagram which shows the structure of a character string upward arrangement | positioning part. It is a block diagram which shows the structure of a character string downward arrangement | positioning part. It is a block diagram which shows the structure of a determination process part. It is a block diagram which shows the structure of the minimum cost function value calculation part. 4 is a flowchart showing a flow of operations performed by the character string arrangement device according to the first embodiment. It is a figure which shows the road link and the character string arrangement | positioning start candidate point which the character string 0 and the character string 1 follow. It is a figure which shows the size of the character contained in the character string 0 and the character string 1, respectively. 6 is a flowchart showing a flow of creating a character string arrangement start candidate point by the character string arrangement apparatus of the first embodiment. It is a figure which shows the road link along the character string 0, and the character string arrangement | positioning start candidate point on the said road link. It is a flowchart which shows the flow of the process which arrange | positions a character string to a character string arrangement | positioning start candidate point, and determines whether arrangement | positioning is possible. It is a figure explaining a character arrangement candidate point. It is a figure explaining the character arrangement order. It is a figure explaining the temporary arrangement | positioning of the character to character arrangement | positioning candidate point P '. It is a flowchart which shows the flow of the determination process of the character arrangement | positioning propriety by the determination process part. It is a figure explaining the temporary arrangement | positioning of the character to character arrangement | positioning candidate point P '. It is a figure which shows the case where the 1st character with arrangement order m is provisionally arranged following the 0th character with arrangement order m. It is a figure which shows an example of temporary arrangement | positioning of the 2nd character whose arrangement order m is. It is a figure which shows another example of temporary arrangement | positioning of the 2nd character whose arrangement order m is. It is a figure which shows another example of temporary arrangement | positioning of the 2nd character whose arrangement order m is. It is a figure which shows an example of upward arrangement of the character string 0. FIG. It is a figure which shows an example of the upward arrangement | positioning and downward arrangement | positioning of the character string 0. FIG. It is a figure which shows an example of the downward arrangement | positioning which cannot be arrange | positioned. It is a figure which shows the character string arrangement | positioning start candidate point of the character string 0 and the character string 1. FIG. It is a flowchart which shows the flow of a calculation process of the character string arrangement | sequence where the cost function value becomes the minimum. It is a flowchart which shows the flow of the calculation process of the duplication function value between character strings. It is a figure which shows the case where the character string 0 and the character string 1 are arrange | positioned upward and downward. It is a flowchart which shows the flow of the calculation process of the function value which shows the angle of the road link which a character string follows. It is a figure which shows an example of a cost function value. It is a block diagram which shows the structure of the character string arrangement | positioning apparatus by Embodiment 2 of this invention. It is a block diagram which shows the structure of the cost function value calculation part for every individual. It is a block diagram which shows the structure of an individual re-creation part. It is a block diagram which shows the structure of a crossing part. It is a figure which shows an example of an individual | organism | solid. 12 is a flowchart showing a flow of operations performed by the character string arrangement device according to the second embodiment. It is a figure which shows an example of the relationship of the road link which the character string 0 and the character string 1 follow, a character string arrangement | positioning start point, and a symbol position. It is a figure which shows an example of initialization of an individual | organism | solid. It is a flowchart which shows the flow of the calculation process of the cost function value for every individual. It is a figure which shows an example of the cost function value in the initialized individual. It is a flowchart which shows the flow of the calculation process of the cost function value which shows duplication of a character and a symbol. It is a figure explaining the creation process of the circumscribed rectangle of a character string. It is a figure explaining the re-creation process of a solid. It is a flowchart which shows the flow of a crossover process. It is a figure explaining a crossover process. It is a figure explaining a mutation process. It is a figure which shows an example of the cost function value for every individual including the recreated individual.

Hereinafter, in order to describe the present invention in more detail, modes for carrying out the present invention will be described with reference to the accompanying drawings.
Embodiment 1 FIG.
FIG. 10 is a block diagram showing the configuration of the character string arrangement device according to Embodiment 1 of the present invention. The character string arrangement device 1 according to the first embodiment is a device that constitutes a map display device used for, for example, a navigation device, and shows information such as a name along a link representing a road on the map. Place a string. In the following description, a case will be described in which the character string placement device 1 places a character string indicating a road name or the like along a road link of the road.
As shown in FIG. 10, the character string arrangement device 1 has a character string data acquisition unit 2, a character string arrangement start candidate point creation unit 3, a minimum cost function value calculation unit 4, and an output character string arrangement start candidate point storage. Part 5 is provided.

The character string data acquisition unit 2 is a component that acquires character string data from an external storage device. The character string data is, for example, a character string of a road name that is included in map data used for map display and displayed on the map screen, road link data indicating a road link along the character string, and characters on the map screen. This is data including a character string arrangement start point indicating a position where arrangement of the column is started.
The character string arrangement start candidate point creation unit 3 starts arrangement of the character string on the map screen when the character string arrangement start point of the character string acquired from the character string data by the character string data acquisition unit 2 is changed. This is a component that creates point candidates (hereinafter referred to as character string arrangement start candidate points).

The minimum cost function value calculation unit 4 uses the character string arrangement start candidate point created by the character string arrangement start candidate point creation unit 3 as a starting point, and when the corresponding character string is arranged upward and downward, As factors that affect readability, functions that indicate the overlap between character strings in each layout, functions that indicate the angle between adjacent road links that cause deformation display of character strings, and characters that are set in advance in character string data A component that calculates a cost function value obtained by adding the values of functions indicating the distance between the column arrangement start point and the character string arrangement start candidate point, and obtains a character string arrangement start candidate point that minimizes the cost function value. .
In the present invention, on a screen with a certain map orientation (one direction on the display screen), the layout is such that the character string can be read to the left while it is divided on the road link along the target character string. Is placed upward. In addition, when the orientation of the map on the screen in the upward orientation is reversed (direction opposite to one direction on the display screen), the character string can be read to the left on the other side separated from the road link. The orientation arrangement is the downward arrangement.

The output character string arrangement start candidate point storage unit 5 is a storage unit that stores a character string arrangement start candidate point at which the cost function value calculated by the minimum cost function value calculation unit 4 is minimized.
The character string arrangement device 1 uses the character string arrangement start candidate point stored in the output character string arrangement start candidate point storage unit 5 to respond to the overlap of character strings or the curve of the road in any map orientation. The character string can be arranged at a position where the character string is not deformed.

  Note that the character string data acquisition unit 2, the character string arrangement start candidate point creation unit 3, and the minimum cost function value calculation unit 4, for example, cause the computer to execute a character string arrangement program according to the gist of the present invention, thereby It is realized as a specific means in which the hardware and software cooperate. That is, each function of the components 2 to 4 is defined in each program module constituting the character string arrangement program. Moreover, the output character string arrangement | positioning start candidate point memory | storage part 5 is constructed | assembled on the storage area of the memory | storage device mounted in said computer, for example. Examples of the storage device include a hard disk device, a semiconductor memory, a drive device for storage media such as a CD and a DVD, and the like.

FIG. 11 is a block diagram showing the configuration of the character string arrangement start candidate point creation unit in FIG. In FIG. 11, the character string arrangement start candidate point creation unit 3 is a component that creates character string arrangement start candidate points for the character string acquired by the character string data acquisition unit 2. An upward placement unit 7, a character string downward placement unit 8, and a candidate point deletion unit 9 are provided.
The candidate point creation unit 6 is a configuration unit that creates a character string arrangement start candidate point of the character string based on the character string acquired from the character string data by the character string data acquisition unit 2 and information on the road link. For example, the position of a predetermined interval on one or a plurality of continuous road links along which the character string extends is set as a character string arrangement start candidate point of the character string.

The character string upward arrangement unit 7 temporarily arranges the character string upward along the road link starting from the character string arrangement start candidate point created by the candidate point creation unit 6, and the character and road link overlap. This is a configuration unit that determines whether or not the arrangement interval is equal to or greater than a predetermined value and whether the characters in the character string overlap each other, and determines whether or not the character string can be arranged upward from this determination result.
The character string downward arrangement unit 8 temporarily arranges the character string downward along the road link starting from the character string arrangement start candidate point created by the candidate point creation unit 6, and the character and road link overlap. This is a configuration unit that determines whether the arrangement interval is equal to or greater than a predetermined value and whether the characters in the character string overlap each other, and determines whether or not the character string can be arranged downward from this determination result.
The candidate point deletion unit 9 uses the character string upward placement unit 7 and the character string downward placement unit 8 among the character string placement start candidate points created by the candidate point creation unit 6 to perform the character placement either upward or downward. This is a configuration unit for deleting a character string arrangement start candidate point where it is determined that a string is not arranged.

FIG. 12 is a block diagram showing the configuration of the character string upward arrangement portion of FIG. In FIG. 12, a character string upward arrangement unit 7 includes a character arrangement candidate point creation unit 10a, a start character arrangement candidate point determination unit 11a, an upward character arrangement order determination unit 12a, an upward character temporary arrangement unit 13a, and a determination processing unit 14a. And a character position registration unit 15a.
The character placement candidate point creation unit 10a uses the character string placement start candidate point created by the candidate point creation unit 6 as a starting point to position each character of the character string to be placed upward (hereinafter referred to as a character placement candidate point). ). For example, a position arranged along a road link at a predetermined interval from a character string arrangement start candidate point is set as a character arrangement candidate point.

The starting character placement candidate point determination unit 11a searches for a character placement candidate point serving as a starting point for character placement of a character string to be placed upward from the character placement candidate points created by the character placement candidate point creation unit 10a. This is a component to be determined.
The upward character arrangement order determination unit 12a determines the character based on the declination of the unit vector of the road link where the character arrangement candidate point serving as the character arrangement start point determined by the start character arrangement candidate point determination unit 11a exists. It is a component which determines whether the character arrangement order of a column will be either a forward direction or a reverse direction. Here, the forward direction is an arrangement order of characters that can read the character string to the left, and the backward direction is an arrangement order that makes the reading of the character string read to the right.

The upward character temporary placement unit 13a starts the character placement determined by the start character placement candidate point determination unit 11a for each character of the character string in the character placement order determined by the upward character placement order determination unit 12a. This is a component that temporarily arranges each character arrangement candidate point in order from the point.
The determination processing unit 14a is configured such that the character temporarily arranged by the upward character temporary arrangement unit 13a overlaps the character and the road link, and the character and the character temporarily arranged immediately before the character are arranged at a predetermined interval or more. It is a configuration unit that determines whether there is an overlap between the character and the character temporarily arranged immediately before the character and determines whether the character can be arranged based on the determination result.
The character position registration unit 15a is a storage unit that registers the coordinate information of the position of the character that is determined to be arrangeable by the determination processing unit 14a as the character position.

FIG. 13 is a block diagram showing the configuration of the character string downward arrangement portion of FIG. In FIG. 13, the character string downward arrangement unit 8 includes a character arrangement candidate point creation unit 10b, a start character arrangement candidate point determination unit 11b, a downward character arrangement order determination unit 16b, a downward character temporary arrangement unit 17b, and a determination processing unit 14b. And a character position registration unit 15b.
The character placement candidate point creation unit 10b is a configuration unit that creates a character placement candidate point for placing each character of the character string to be placed downward from the character string placement start candidate point created by the candidate point creation unit 6. For example, as in the upward character temporary arrangement unit 13a, a position arranged along the road link at a predetermined interval from the character string arrangement start candidate point is set as the character arrangement candidate point.

The starting character placement candidate point determination unit 11b searches for a character placement candidate point serving as a starting point for character placement of a character string to be placed downward from the character placement candidate points created by the character placement candidate point creation unit 10b. This is a component to be determined.
The downward character arrangement order determination unit 16b determines the character based on the declination of the unit vector of the road link where the character arrangement candidate point serving as the character arrangement start point determined by the start character arrangement candidate point determination unit 11b exists. It is a component which determines whether the character arrangement order of a column will be either a forward direction or a reverse direction.

The downward character temporary arrangement unit 17b starts the character arrangement determined by the start character arrangement candidate point determination unit 11b for each character of the character string in the character arrangement order determined by the downward character arrangement order determination unit 16b. This is a component that temporarily arranges each character arrangement candidate point in order from the point.
For the character temporarily arranged by the downward character temporary arrangement unit 17b, the determination processing unit 14b overlaps the character and the road link, and the character and the character temporarily arranged immediately before the character are arranged at an arrangement interval of a predetermined value or more. It is a configuration unit that determines whether there is an overlap between the character and the character temporarily arranged immediately before the character and determines whether the character can be arranged based on the determination result.
The character position registration unit 15b is a storage unit that registers, as a character position, coordinate information of a character position that is determined to be arrangeable by the determination processing unit 14b.

FIG. 14 is a block diagram illustrating a configuration of the determination processing unit illustrated in FIG. 12 or 13. In FIG. 14, the determination processing units 14 a and 14 b include a character / road overlap determination unit 18, a character interval determination unit 19, and a character overlap determination unit 20.
The character and road duplication determination unit 18 of the determination processing unit 14a is a configuration unit that determines whether or not there is an overlap with a road link along which a character string including the character along the character temporarily arranged in the upward character temporary arrangement unit 13a. is there. In the character and road duplication determination unit 18 of the determination processing unit 14b, a component that determines whether or not there is duplication with a road link along which a character string including the character is temporarily arranged with respect to the character temporarily arranged in the downward character temporary arrangement unit 17b. is there.

The character interval determination unit 19 of the determination processing unit 14a determines whether the interval between the character temporarily arranged by the upward character temporary arrangement unit 13a and the character temporarily arranged immediately before is narrower or wider than a predetermined character interval. It is the component which determines. In addition, the character interval determination unit 19 of the determination processing unit 14b is configured such that the interval between the character temporarily arranged in the downward character temporary arrangement unit 17b and the character temporarily arranged immediately before is narrower or wider than a predetermined character interval. This is a component that determines whether or not.
The character overlap determination unit 20 in the determination processing unit 14a is a component that determines the overlap between the character temporarily arranged in the upward character temporary arrangement unit 13a and the character temporarily arranged immediately before. The character overlap determination unit 20 of the determination processing unit 14b is a component that determines the overlap between the character temporarily arranged by the downward character temporary arrangement unit 17b and the character temporarily arranged immediately before.

15 is a block diagram showing the configuration of the minimum cost function value calculation unit in FIG. 10, FIG. 15 (a) shows the configuration of the minimum cost function value calculation unit, and FIG. 15 (b) shows the configuration of FIG. The structure of the duplication function calculation part of the character strings in (a) is shown.
In FIG. 15A, the minimum cost function value calculation unit 4 includes a duplication function calculation unit 21 between character strings, a road link angle function calculation unit 22, a distance function calculation unit 23, and a function value addition unit 24. As shown in FIG. 15B, the duplication function calculation unit 21 between character strings includes a character string upward arrangement unit 25a, a character string downward arrangement unit 25b, and a character string duplication calculation unit 25c.

The character string duplication function calculation unit 21 starts the character string arrangement start determined by the character string arrangement start candidate point creation unit 3 for each character string acquired from the character string data by the character string data acquisition unit 2. This is a configuration unit that determines the overlap between character strings when the candidate points are arranged as starting points, using a function indicating the overlap between the character strings.
The road link angle function calculation unit 22 is a component that calculates a value of a function Angle (i) that indicates an angle formed by a road link along which a character string acquired from character string data follows.
The distance function calculation unit 23 includes the character string arrangement start point of the character string acquired from the character string data by the character string data acquisition unit 2 and the character string arrangement determined to be arrangeable by the character string arrangement start candidate point creation unit 3 It is a component that calculates a value of a function Dist (i) indicating a distance from a start candidate point.

The character string upward arrangement unit 25a determines the character string arrangement start candidate point determined by the character string arrangement start candidate point creation unit 3 for each character string acquired from the character string data by the character string data acquisition unit 2. It is a component arranged upward as a starting point.
In addition, the character string downward arrangement unit 25b has the character string arrangement start candidate determined by the character string arrangement start candidate point creation unit 3 for each character string acquired from the character string data by the character string data acquisition unit 2. It is a component which arranges downwards from a point as a starting point.
The character string duplication calculation unit 25c includes the duplication of character strings arranged upward by the character string upward arrangement unit 25a, the duplication of character strings arranged downward by the character string downward arrangement unit 25b, and the character string upward arrangement unit 25a. The value of the function Over _String (i) indicating the duplication of the character strings is obtained by using the duplication of the character strings obtained by the duplication judgment between the character strings arranged upward and downward by the character string downward arrangement unit 25b. It is a component to calculate.

Function value adding mating portion 24, the function _{Overlap String} (i), the value of the function Angle (i) and the function Dist (i) the sum total of consisting cost function f (i), the string with the string data acquisition unit 2 The character string arrangement start point of the character string acquired from the data and the character string arranged at the character string arrangement start candidate point determined to be arrangeable by the character string arrangement start candidate point creation unit 3 are calculated, and the cost function f This is a component that sets the point at which the value of (i) is the smallest as the final character string arrangement start candidate point. The final character string arrangement start candidate point storage unit 5 stores the final character string arrangement start candidate point.

Next, the operation will be described.
FIG. 16 is a flowchart showing a flow of operations performed by the character string arrangement device according to the first embodiment. The outline of the process of acquiring the character string arrangement start candidate point will be described with reference to FIG.
First, the character string data acquisition unit 2 acquires character string data from an external storage device such as a map database, and all character strings included in the character string data, and road links and character strings corresponding to the respective character strings. An arrangement start point is extracted (step ST1).
Next, the character string arrangement start candidate point creation unit 3 creates character string arrangement start candidate points on the road links along which the respective character strings acquired by the character string data acquisition unit 2 follow (step ST2). At this time, the character string arrangement start candidate point creation unit 3 determines whether or not the character string arrangement start candidate point can be arranged for each character of each character string.
Thereafter, the minimum cost function value calculation unit 4 creates the character string arrangement start candidate point creation unit 3 and determines that the character string arrangement start candidate points can be arranged. The angle formed by the road link along which the character string along which the character string is displayed and the distance indicating the degree of difference between the character string arrangement start candidate point and the character string arrangement start point preset in the character string data, respectively A cost function including the function shown is calculated, and a character string arrangement start candidate point (character string position) that minimizes the cost function value is acquired (step ST3).

In the following, it is assumed that the character string data acquisition unit 2 has acquired two character strings of character string 0 and character string 1 from the character string data, and character string 0 and character string 1 and road links corresponding to these character strings. The character string arrangement start point is assumed to be defined as follows.
FIG. 17 is a diagram showing a road link and a character string arrangement start candidate point along which character string 0 and character string 1 respectively follow. In a map screen in which the map orientation is north (north up), the east-west direction is the x-axis, A two-dimensional coordinate system is set with the north-south direction as the y-axis.
The road links along the character string 0 are the links of the nodes A (0, 20), B (10, 20), C (10, 10), and D (30, 10), respectively (broken lines in FIG. 17). Link). In addition, it is assumed that the character string arrangement start point of the character string 0 preset in the character string data is I (12, 10).
Further, the road links along the character string 1 connect the node E (30, 0), the node D (30, 10), and the node F (30, 30) with links (links indicated by solid lines in FIG. 17). Consists of. It is assumed that the character string arrangement start point of the character string 1 preset in the character string data is J (30, 12).

FIG. 18 is a diagram showing the sizes of characters included in the character string 0 and the character string 1, respectively. FIG. 18 (a) shows the character size of the character string 0, and FIG. The size of 1 character is shown. As shown in FIG. 18A, each of the characters “sentence”, “character”, “column”, and “0” included in the character string 0 has a vertical width of 5 and a horizontal width of 3. Further, as shown in FIG. 18B, the “sentence”, “character”, “column”, and “1” characters included in the character string 1 have a vertical width of 5 and a horizontal width of 3 similarly. It shall be.
It is assumed that the distance RCgap between the road link and each character of the character string 0 and the character string 1 is “3”, and the character-to-character distance CCgap is “√2 (square root of 2)”.

The character string arrangement start candidate point creation unit 3 creates five character string arrangement start candidate points PATTERN_NUM for each character string, and the character arrangement candidate point interval DIF_CAND_POINT for arranging each character of the character string. There "2" is assumed to be, the threshold _overlap string threshold of duplicate numbers between string is assumed to be "5".

Further, all the parameters α ₁ , α ₂ , α ₃ for calculating cost function values described later are set to “1”. The maximum value of the above-mentioned intervals RCgap, CCgap, DIF_CAND_POINT, etc. that can be processed by the computer functioning as the character string arranging device 1 is “MAX”.
In addition, the value of each parameter mentioned above shall be preset for every character string in character string data.

Next, details of the process of acquiring the character string arrangement start candidate point will be described.
(1) Character String Arrangement Start Candidate Point Creation Process FIG. 19 is a flowchart showing a flow of creating a character string arrangement start candidate point by the character string arrangement apparatus according to the first embodiment, and the process of step ST2 in FIG. Details are shown. Refer to FIG. 11 for the configuration of the character string arrangement start candidate point creation unit 3.
First, the candidate point creation unit 6 inputs all the character strings included in the character string data, road links and character string arrangement start points corresponding to these character strings from the character string data acquisition unit 2. Here, each character string input from the character string data acquisition unit 2 is represented by a character string (i) having an index number i that can take a value from 0 to the number of character strings included in the character string data minus one.
Candidate point creation unit 6 performs the subsequent processing in order from the character string (0) of i = 0 among the character strings input from character string data acquisition unit 2 (step ST1a).

The candidate point creation unit 6 creates character string arrangement start candidate points for the number of candidate points PATTERN_NUM (here, 5) of the character string (i) in the road link along the character string (i) (step ST2a). That is, the candidate point creation unit 6 acquires the full length LinkLength of the road link from the data related to the road link along which the character string (i) follows, and sets the distance CandDif of the character string arrangement start candidate points of the character string (i) from the following formula: calculate.
CandDif = LinkLength / PATTERN_NUM

For the character string 0, if the total length LinkLength of the road links where the nodes A to D shown in FIG. CandDif = 40/5 = 8.
Subsequently, the candidate point creation unit 6 creates a character string arrangement start candidate point from the first node of the road link along the character string (i) by the number of candidate points PATTERN_NUM at the interval CandDif along the road link.
The first node of the road link along the character string (i) can be a character string arrangement start candidate point, but the end node of the road link along the character string (i) is a character string arrangement start candidate point. Do not become.
In this way, as shown in FIG. 20, the character string arrangement start candidate points Q0 (0, 20), Q1 (8, 20), Q2 (10, 14), Q3 (14, 10), and Q4 (22, 10) are created.
Each character string arrangement start candidate point is represented by a character string arrangement start candidate point (j) having an index number j that can take a value from 0 to the number of candidate points PATTERN_NUM.
That is, the character string arrangement start candidate point (0) is Q0, the character string arrangement start candidate point (1) is Q1, the character string arrangement start candidate point (2) is Q2, the character string arrangement start candidate point (3) is Q3, The character string arrangement start candidate point (4) is Q4.
Candidate point creation unit 6 performs processing in order from the character string arrangement start candidate point (0) of j = 0 among the character string arrangement start candidate points created in step ST2a (step ST3a).

Next, the character string upward arrangement unit 7 arranges the character string (i) upward from the character string arrangement start candidate point (j), and determines whether or not the character string (i) can be arranged (step ST4a).
Subsequently, the character string downward arrangement unit 8 arranges the character string (i) downward from the character string arrangement start candidate point (j), and determines whether or not the character string (i) can be arranged (step ST5a). Details of these processes will be described later. Moreover, although the case where the arrangement | positioning availability determination by upward arrangement | positioning is performed previously was shown, you may make it perform the arrangement | positioning availability determination by downward arrangement | positioning first.

The candidate point deletion unit 9 determines whether the character string arrangement start candidate point (j) can be arranged in a character string based on the determination result of the character string arrangement possibility by the character string upward arrangement unit 7 and the character string downward arrangement unit 8. Is determined (step ST6a). Here, it is determined that the character string cannot be arranged either when the character string (i) is arranged upward by the character string upward arrangement unit 7 or when the character string (i) is arranged downward by the character string downward arrangement unit 8. In this case, this character string arrangement start candidate point (j) cannot be arranged in the character string.
When the character string arrangement is impossible (step ST6a; NO), the candidate point deletion unit 9 deletes the character string arrangement start candidate point (j) (step ST7a). If the character string can be arranged (step ST6a; YES), the character string arrangement start candidate point (j) is left, and the process proceeds to step ST8a.

In step ST8a, the candidate point creation unit 6 determines whether or not the processing after step ST4a has been performed for all the character string arrangement start candidate points (j).
Here, if the index number j is less than the number of character string arrangement start candidate points (PATTERN_NUM = 5) and there is a character string arrangement start candidate point (j) for which the above processing has not been performed (step ST8a; YES), candidates The point creation unit 6 outputs the character string arrangement start candidate point (j) obtained by adding +1 to the index number j to the character string upward arrangement unit 7 (step ST9a). Thereby, the process after step ST4a is implemented about the subsequent character string arrangement | positioning start candidate point (j).

On the other hand, when the value of the parameter j is equal to or greater than the number of character string arrangement start candidate points (PATTERN_NUM = 5) and there is no character string arrangement start candidate point (j) for which the above processing has not been performed (step ST8a; NO), candidate points The creation unit 6 determines whether or not the processing after step ST2a has been performed for all character strings (i) (step ST10a).
If the index number i is less than the number of character strings and there is a character string (i) for which the above processing has not been performed (step ST10a; YES), the candidate point creation unit 6 adds a character string obtained by adding +1 to the index number i. (I) is specified (step ST11a), and the processing after step ST2a is performed. If the processing has been completed for all the character strings (i) (step ST10a; NO), the processing ends.

Next, details of the character string arrangement and arrangement availability determination processing by the character string upward arrangement unit 7 and the character string downward arrangement unit 8 will be described.
FIG. 21 is a flowchart showing a flow of processing for determining whether or not to arrange a character string at a character string arrangement start candidate point, and corresponds to the processing of step ST4a and step ST5a in FIG. That is, both the character string upward arrangement unit 7 and the character string downward arrangement unit 8 determine the character string arrangement and the arrangement possibility in the process flow shown in FIG. Refer to FIGS. 7 and 8 for the configuration of the character string upward arrangement unit 7 and the character string downward arrangement unit 8.

(1-1) Upward Arrangement of Character String and Determination of Arrangeability A case where the character string upward arrangement unit 7 arranges the character string 0 upward starting from the character string arrangement start candidate point Q0 will be described with reference to FIG.
First, the character placement candidate point creation unit 10a of the character string upward placement unit 7 creates character placement candidate points sequentially from the node A shown in FIG. 17 at an interval DIF_CAND_POINT (= 2) (step ST1b). For example, as shown in FIG. 22A, character placement candidate points P0 to P20 are created on the road link. FIG. 22B shows the coordinate values of the character arrangement candidate points P0 to P20. Also, each character placement candidate point created in the character placement candidate point creation unit 10a is represented by a character placement candidate point (k) with an index number k that can take a value from 0 to the number of character placement candidate points−1. . That is, the character arrangement candidate points P0 to P20 correspond to the index numbers k = 0 to 20, respectively, and are represented by the character arrangement candidate points (0) to the character arrangement candidate points (20).

  Next, the starting character placement candidate point determination unit 11a searches for a character placement candidate point P ′ at which the character placement is started (step ST2b). The character arrangement candidate point P ′ is a point having the smallest distance from the character string arrangement start candidate point Q0 among the character arrangement candidate points P0 to P20. Here, from FIG. 20 and FIG. 22A, the character arrangement candidate point P ′ is the character arrangement candidate point P0 (character arrangement candidate point (0)).

Subsequently, the upward character arrangement order determination unit 12a specifies a link where the character arrangement candidate point P ′ is present among the links forming the road link along the character string 0, and the unit vector LoadVec (cos θ, Using the declination angle θ of sin θ), it is determined from the following formula (2) whether the character arrangement order is the forward direction or the reverse direction (step ST3b).
Here, the point P0 which is the character arrangement candidate point P ′ exists in a link connecting the node A and the node B (hereinafter referred to as a road link {A, B}), so the unit vector is (cos0, sin0). ) Therefore, the character arrangement order is forward.
Note that the forward and reverse character arrangement order here refers to the direction in which the characters in the character string can be read to the left, as shown in FIGS. 23 (a) and 23 (b). The direction for right reading is the reverse direction. The character arrangement order is expressed by an index number m that can take values from 0 to the number of characters −1 in the character string (i).
if −π / 2 <θ ≦ π / 2 Forward direction else Reverse direction (2)

When the character arrangement order is determined, the upward character temporary arrangement unit 13a uses the character string 0 and the arrangement order m = 0 for the character arrangement candidate point (character arrangement candidate point P ′, k = 0) to start character arrangement. The following processing is performed in order from the character (step ST4b).
The upward character temporary placement unit 13a temporarily places the m-th character in the placement order at the character placement candidate point P ′ using the following equation (3) (step ST5b).
Where P _xk and P _yk indicate the x-coordinate and y-coordinate of the k-th character placement candidate point, CharVec (cosφ, sinφ) indicates a unit vector from the road link to the character, and char_height _m is The vertical width of the mth character, char_width _m , indicates the horizontal width of the mth character.
Also, (x _k ^BottomLeft , y _k ^BottomLeft ) in the following formula (3) indicates the lower left coordinates of the character, and (x _km ^TopLeft , y _km ^TopLeft ) indicates the upper left coordinates of the character.
Further, (x _km ^BottomRight , y _km ^BottomRight ) indicates the lower right coordinate of the character, and (x _km ^TopRight , y _km ^TopRight ) indicates the upper right coordinate of the character.

FIG. 24 is a diagram for explaining the temporary arrangement of characters at the character arrangement candidate point P ′. When the character “sentence” (forward direction) having the youngest arrangement order in the character string 0 is temporarily arranged at the character arrangement candidate point P ′ (character arrangement candidate point P0), the lower left coordinates and the upper left coordinates represented by the above formula (3) The lower right coordinates and upper right coordinates are as shown in FIG. The broken-line rectangle is a circumscribed rectangle of the character “sentence”.
In FIG. 24, the unit vector CharVec (cosφ, sinφ) from the road link to the character is calculated from the following equation (4) using the deflection angle θ of LoadVec (cosθ, sinθ).
if −π / 2 ≦ θ <π / 2 φ = θ−π / 2
if π / 2 ≦ θ <3π / 2 φ = θ + π / 2 (4)

  Next, the determination processing unit 14a determines whether or not the character readability is impaired at the position temporarily arranged in the upward character temporary arrangement unit 13a (step ST6b), and based on the determination result. It is determined whether or not the character arrangement at the temporary arrangement position is possible (step ST7b). Here, if the determination processing unit 14a determines that the character arrangement is possible (step ST7b; YES), the character arrangement position is registered in the character position registration unit 15a (step ST8b). At this time, the four vertices of the lower left coordinate, the upper left coordinate, the lower right coordinate, and the upper right coordinate of the circumscribed rectangle of the character are registered as the character position.

Next, the upward character temporary arrangement unit 13a determines whether or not the above-described processing has been performed for all characters of the character string (i) (here, the character string 0) (step ST9b).
Here, when the value of the character arrangement order m is equal to the number of characters -1 in the character string 0 and the processing is performed for all characters (step ST9b; YES), the process proceeds to step ST10b. In step ST10b, since the determination processing unit 14a can arrange all the characters in the character string 0, it determines that the character string 0 can be arranged and ends the processing.

  When the value of the character arrangement order m does not reach the number of characters −1 in the character string 0 and it is determined that there is an unprocessed character (step ST9b; NO), the upward character temporary arrangement unit 13a determines the value of the index number k. Does not reach the number of character placement candidate points, and there is a character placement candidate point (k) for which the above processing has not been performed, the index number k and index number m are respectively incremented by +1, and the character to be processed next (m = 1) is temporarily arranged at the next character arrangement candidate point (1) (step ST11b). Thereby, the process after step ST6b is implemented with respect to this arrangement | positioning.

  On the other hand, when the determination processing unit 14a determines that the character cannot be placed (step ST7b; NO), the upward character temporary placement unit 13a does not reach the number of character placement candidate points because the index number k does not reach the number of character placement candidate points. If there is an unexecuted character placement candidate point (k), the character is temporarily placed at the next character placement candidate point (k), and the process returns to step ST6b. If the processing is completed for all the character placement candidate points (k), it is determined that the placement of the character string 0 is impossible and the processing is terminated. These processes correspond to step ST12b.

(1-2) Determination Processing FIG. 25 is a flowchart showing the flow of determination processing for determining whether or not a character can be placed by the determination processing section, and shows details of the processing in step ST6b in FIG. Refer to FIG. 14 for the configuration of the determination processing unit 14a.
First, when a character is temporarily arranged by the upward character temporary arrangement unit 13a, the character and road duplication determination unit 18 searches for an overlapping portion of the character and the road at the temporarily arranged position, and determines whether or not there is an overlap. Determination is made (step ST1c). In the example shown in FIG. 20, the circumscribed rectangle of the character “sentence” whose arrangement order m of the character string 0 is 0 and the road link {A, B, C, D} that is the road link along the character string 0 overlap. Is searched.

FIG. 26 is a flowchart showing the flow of determination of overlap between characters and roads, and shows details of the processing in step ST1c in FIG.
First, each link of a road link to be subjected to duplication determination is expressed by a link (n) having an index number n that can take a value from 0 to the number of links of the road link. The character and road duplication determination unit 18 performs processing in order from the link (0) of n = 0 among the links of the road links along which the character string of the character temporarily arranged by the upward character temporary arrangement unit 13a follows ( Step ST1d).
The character and road duplication determination unit 18 searches for an overlap between the character temporarily arranged by the upward character temporary arrangement unit 13a and the link (n) (step ST2d), and determines the presence or absence of duplication from the search result (step ST2d). Step ST3d). Here, if there is no overlap, the character / road overlap determination unit 18 stores the fact as determination result data (step ST4d), and if there is a overlap, stores that fact as determination result data (step ST5d).

  After that, the character and road duplication determination unit 18 determines whether or not duplication determination with all links of the road links to be subjected to duplication determination has been performed (step ST6d). Here, when the value of the link index number n does not reach the number of links of the road link to be duplicated and the processing has not been performed for all the links (step ST6d; YES), the character and road duplication judgment unit 18 adds +1 to the index number n, specifies the link (n = 1) to be processed next (step ST7d), and performs the processing after step ST2d. Further, when the value of the index number n of the links reaches the number of links of the road link to be determined for duplication, and the processing is performed for all the links (step ST6d; NO), the processing is terminated.

In the example shown in FIG. 20, first, of the road links {A, B, C, D} along which the character string 0 follows, the overlap between the link {A, B} and the character “sentence” is determined. It is determined that the link {A, B} and the character “sentence” do not overlap. Next, the overlap between the link {B, C} and the character “sentence” is determined. When it is determined that the road link {A, B, C, D} overlaps with the character “sentence” by repeating the same process, the character “sentence” overlaps with the road link {A, B, C, D}. I understand that I do not. For this reason, it is determined that there is no overlap by the character / road overlap determination unit 18.
If there is an overlap between the character and the road, it is determined that the character cannot be placed at the temporarily placed position. If there is no overlap, the process proceeds to the next character interval determination process.

Returning to the description of FIG.
The character interval determination unit 19 calculates the intervals between the characters temporarily arranged by the upward character temporary arrangement unit 13a and all the characters arranged immediately before, and compares the character intervals with a predetermined threshold value. Whether the character spacing is wide or narrow is determined (step ST2c).
In the calculation method of the character spacing, first, the distances between the vertices of the four vertices of the circumscribed rectangle of the temporarily arranged character and the four vertices of the circumscribed rectangle of all the characters arranged immediately before are respectively determined. Thereafter, the smallest distance among the obtained distances between the vertices is set as the character spacing between the characters.
At this time, if the temporarily placed character is the 0th character in the placement order m, there is no character placed before that, so the character spacing is set to MAX. That is, in the character string 0, since there is no character arranged before the character “sentence”, the character interval is MAX.
Next, if the character spacing is greater than or equal to CCgap, it is determined that the character spacing is wide, and if it is less than CCgap, it is determined that the character spacing is narrow. The character “sentence” of the character string 0 is “character wide” because the character spacing is MAX. Note that the CCgap serving as a threshold is a value preset in the character string data as described above.
If the character spacing is “narrow character spacing”, it is determined that the character cannot be placed at the temporarily placed position.

The character duplication determination unit 20 searches for an overlapping portion between the character temporarily arranged by the upward character temporary arrangement unit 13a and all the characters arranged immediately before, and determines the presence or absence of duplication ( Step ST3c). Here, if it is determined that there is an overlap between the characters, it is determined that the character cannot be placed at the temporarily placed position. If there is no overlap, the process ends.
If the temporarily placed character is the 0th character in the placement order m, there is no character placed before that, so there is no overlap between the characters. For this reason, it is determined that there is no overlap between characters.

When the character duplication determination unit 20 finally determines that there is no character duplication as described above, the character position registration unit 15a registers the temporarily arranged character position.
In the case of the character “sentence” of the character string 0, since it is determined that the character placement is possible at the character placement candidate point P ′, the character position registration unit 15a stores the lower left, upper left, lower right, The coordinates of the four vertices in the upper right are registered as character positions. That is, as shown in FIG. 24, the character position registration unit 15a stores the lower left coordinates (0, 17), upper left coordinates (0, 12), lower right coordinates (3, 17), upper right coordinates ( 3, 12) is registered.

Thereafter, the upward character temporary arrangement unit 13a temporarily arranges the first character “character” in the arrangement order m at the next character arrangement candidate point P1.
FIG. 27 is a diagram illustrating a case where the first character in the arrangement order m is provisionally arranged following the character in the arrangement order m of the 0th character. In FIG. 27, the circumscribed rectangle with the symbol A is the arrangement position of the character “sentence” whose arrangement order m is the 0th character, and the circumscribed rectangle with the symbol B is the character “character” with the first arrangement order m. Is the arrangement position. The character “sentence” whose arrangement order m is 0th is arranged based on the character arrangement candidate point P ′ (character arrangement candidate point P0) at which character arrangement starts, but the first character “character” whose arrangement order m is 1 Are temporarily arranged based on the character arrangement candidate point P1 next to the character arrangement candidate point P ′.

When the next character in the character string 0 is temporarily arranged, the character / road overlap determination unit 18 determines the overlap between the character and the road in the same manner as described above. In this case, it is determined that the character “letter” does not overlap with the road.
Thereafter, the character spacing determining unit 19 obtains the character spacing between the temporarily placed character “character” and the immediately preceding character “sentence”, and determines whether or not the character placement is possible based on the character spacing. To do. In this case, the distances between the four vertices of the circumscribed rectangle A of the character “sentence” and the four vertices of the circumscribed rectangle B of the character “character”, that is, a total of 16 intervertex distances are calculated. Among these inter-vertex distances, the minimum inter-vertex distance is “1”, and since this inter-vertex distance is equal to or less than the threshold CCgap (= √2), it is determined that “character spacing is narrow”.
Therefore, it is determined that it is impossible to arrange the character “character” at the character arrangement candidate point P1.

  If it is determined that the character placement candidate point P1 cannot be placed, the upward character temporary placement unit 13a temporarily places the character “character” at the subsequent character placement candidate point P2, so that the determination processing unit 14a Repeat the determination process. If it is determined that the character “letter” cannot be placed at the character placement candidate point P2, the character “letter” is further temporarily placed at the next character placement candidate point P3.

FIG. 28 is a diagram illustrating an example of a temporary arrangement of the second character in the arrangement order m. In FIG. 28, the circumscribed rectangle attached with the reference symbol A1 is the arrangement position of the character “sentence” with the arrangement order m being the 0th character, and the circumscribed rectangle attached with the reference symbol A2 is the character “character” with the first arrangement order m. Is the arrangement position.
In addition, the circumscribed rectangle to which the symbol B1 is attached is the position where the second character “string” in the arrangement order m is provisionally arranged. In FIG. 28, the circumscribed rectangle B1 of the character “row” is provisionally arranged in parallel with the road link {A, B}, similarly to the circumscribed rectangles A1 and A2, with the character arrangement candidate point P5 as a reference.

  After this, the character / road overlap determination unit 18 determines the overlap between the character “string” and the road link. At this time, as shown in FIG. 28, the circumscribed rectangle B1 of the character “string” and the link {B, C} are searched for overlapping, and it is determined that there is overlapping. Therefore, the character “string” cannot be arranged at the character arrangement candidate point P5.

FIG. 29 is a diagram illustrating another example of the temporary arrangement of the second character in the arrangement order m. In FIG. 29, the circumscribed rectangle A1 is the arrangement position of the character “sentence” whose arrangement order m is the 0th character, and the circumscribed rectangle A2 is the arrangement of the character “character” whose arrangement order m is the first. Position.
In the example shown in FIG. 29, the upward character temporary arrangement unit 13a parallels the circumscribed rectangle B1 of the second character “column” in the arrangement order m to the road link {B, C} with the character arrangement candidate point P5 as a reference. Is temporarily arranged. Also in this case, as shown in FIG. 29, the circumscribed rectangle B1 of the character “string” and the link {A, B} are searched for overlapping, and it is determined that there is overlapping.
Therefore, the character “string” cannot be arranged at the character arrangement candidate point P5.

FIG. 30 is a diagram illustrating still another example of the temporary arrangement of the second character in the arrangement order m. The example shown in FIG. 30 shows a case where the above-described process is repeated and the upward character temporary arrangement unit 13a temporarily arranges the circumscribed rectangle B1 of the character “row” at the character arrangement candidate point P8. In this case, the character / road overlap determining unit 18 determines that there is no overlap between the character and the road.
In addition, the character interval determination unit 19 determines that the character interval (minimum value of the distance between vertices) between the circumscribed rectangle A1 of the character “sentence” and the circumscribed rectangle A2 of the character “character” and the circumscribed rectangle B1 of the character “string” is Since √2 (square root of 2), it is determined that the character gap is greater than the threshold CCgap and the character spacing is wide.
However, the character duplication determination unit 20 searches for the overlap between the circumscribed rectangle B1 of the character “string” and both the circumscribed rectangle A1 of the character “sentence” and the circumscribed rectangle A2 of the character “character”. judge.
Therefore, it is determined that the character “string” cannot be placed at the character placement candidate point P8.

By determining whether or not the character string 0 can be arranged for each character by repeating the above-described processing, it is determined that the character string 0 can be arranged while maintaining readability when arranged upward as shown in FIG.
At this time, the character arrangement candidate point of each character in the upward arrangement maintaining the readability of the character string 0 is determined. In the example of FIG. 31, the character “sentence” is the character placement candidate point P0, the character “letter” is the character placement candidate point P3, the character “string” is the character placement candidate point P10, and the character “0” is the character placement candidate point P11. Deploy. In this case, since all the characters in the character string 0 can be arranged on one side (upward) with the road link as a boundary with respect to the character arrangement candidate points P0 to P20, the character string 0 can be arranged upward.

(1-3) Character String Downward Arrangement and Arrangeability Determination After the character string upward arrangement unit 7 arranges the character string 0 upward and determines whether it can be arranged, the character string downward arrangement unit 8 continues to determine the character string 0. It is determined whether or not it is possible to place it downward.
The flow of the downward arrangement of the character string 0 and the determination of whether or not to arrange the character string 0 by the character string downward arrangement unit 8 is the same as that in FIG.
First, the character placement candidate point creation unit 10b of the character string downward placement unit 8 creates a character placement candidate point. Thereafter, the start character placement candidate point determination unit 11b searches for a character placement candidate point for starting character placement in the same manner as described above. As a result, the character string arrangement candidate points are as shown in FIG. 22A, and the character arrangement candidate point P ′ at which the character arrangement is started becomes the character arrangement candidate point P0.

Next, the downward character arrangement order determination unit 16b determines the character arrangement order from the following equation (5) using the unit vector LoadVec (cos θ, sin θ) of the link where the character arrangement candidate point P ′ exists.
if −π / 2 <θ ≦ π / 2 reverse direction else forward direction (5)

Thereafter, the downward character provisional arrangement unit 17b provisionally arranges the character “0” having the arrangement order m in the character string 0 that is the 0th character with reference to the character arrangement candidate point P0. Note that the vertex coordinates (four vertices) of the circumscribed rectangle of the temporarily arranged character are calculated from the following equation (6).

Φ in the above formula (6) can be calculated by the following formula (7).
if −π / 2 ≦ θ <π / 2 φ = θ + π / 2
if π / 2 ≦ θ <3π / 2 φ = θ−π / 2 (7)

  Next, the determination processing unit 14b determines whether or not the character “0” can be placed at the character placement candidate point P0. Since this determination process is the same as that of the determination processing unit 14a described above, description thereof is omitted. Thereby, the character arrangement candidate point of each character in the downward arrangement maintaining the readability of the character string 0 is determined.

  FIG. 32 is a diagram illustrating an example of the upward arrangement and the downward arrangement of the character string 0. The character string 0 is arranged in the upward and downward directions described above, and the determination of whether or not the character string 0 is arranged is repeated for each character as shown in FIG. In this case, since all the characters in the character string 0 can be arranged on the other side (downward) with the road link as a boundary with respect to the character arrangement candidate points P0 to P20, the character string 0 can be arranged downward. .

(1-4) Deletion Processing of Character String Arrangement Start Candidate Points That Cannot Be Arranged The candidate point deletion unit 9 of the character string arrangement start candidate point creation unit 3 performs the arrangement feasibility determination in the upward arrangement and the downward arrangement as described above. When completed, the character string arrangement start candidate point in which either the character string arrangement in the upward direction or the character arrangement in the downward direction cannot be arranged is deleted from the character string arrangement start candidate points created by the candidate point creation unit 6.
In the example shown in FIG. 20, since the character string 0 can be arranged upward and downward along the road link {A, B, C, D} from the character string arrangement start candidate point Q0, the character string arrangement start candidate point Q0 is Not deleted.
Thereafter, the character string upward arrangement unit 7 and the character string downward arrangement unit 8 arrange the character string 0 upward along the road link {A, B, C, D} with the character string arrangement start candidate point Q1 as a starting point. And it arrange | positions downward and determines whether arrangement | positioning is possible respectively.
Even in this case, since the character string 0 can be arranged upward and downward along the road link {A, B, C, D} from the character string arrangement start candidate point Q1, the character string arrangement start candidate point Q0 is Not deleted.

Next, the character string upward arrangement unit 7 arranges the character string 0 upward along the road link {A, B, C, D} with the character string arrangement start candidate point Q2 as a starting point, and determines whether or not arrangement is possible.
Thereafter, the character string downward arrangement unit 8 determines whether or not the character string 0 can be arranged by arranging the character string 0 downward along the road link {A, B, C, D} starting from the character string arrangement start candidate point Q2.
In this case, by performing the above-described processing, as shown in FIG. 33, the second character “character” in the arrangement sequence m of the character string 0 can be arranged downward. At this time, the character arrangement candidate point with the arrangement character m of the 0th character “0” is the point P8, the character arrangement candidate point with the arrangement sequence m of the first character “string” is the point P15, and the arrangement sequence m is 2. The character arrangement candidate point of the second character “character” is P18.
When the third character “sentence” in the arrangement order m is provisionally placed at the character placement candidate point P20 by the downward character temporary placement unit 17b, the character and road duplication determination unit 18 determines that the character “sentence” and the road Judgment of overlap with. Here, since there is no overlap between the character “sentence” and the road link {A, B, C, D}, it is determined that there is no overlap.
When the overlap between the character and the road is determined, the character interval determination unit 19 determines the character interval. In this case, since the character spacing is “1”, it is less than CCgap and it is determined that the character spacing is narrow. As a result, the character “sentence” cannot be placed at the character placement candidate point P20.
Subsequently, the downward character temporary placement unit 17b tries to temporarily place the character “sentence” at the next character placement candidate point, but there is no next character placement candidate point of P20. For this reason, the downward character temporary placement unit 17b indicates that it is impossible to place the character string 0 along the road link {A, B, C, D} starting from the character string placement start candidate point Q2. The candidate point deletion unit 9 is notified. As a result, the candidate point deletion unit 9 deletes the character string arrangement start candidate point Q2 from the character string arrangement start candidate points created by the candidate point creation unit 6.

  By repeating the above processing, the character string arrangement start candidate points of the character string 0 become Q0 (0, 20) and Q1 (8, 20), and the character string arrangement start candidate points of the character string 1 are shown in FIG. Thus, Q5 (30, 0) and Q6 (30, 6) are obtained. In FIG. 34, links indicated by broken lines are road links {A, B, C, D} along which character string 0 follows, and links indicated by solid lines are road links {E, D, F along which character string 1 extends. } Link.

(2) Evaluation of readability using a cost function When a character string arrangement start candidate point is created by the character string arrangement start candidate point creation unit 3 as described above, the minimum cost function value calculation unit 4 calculates the following formula: Using the cost function f represented by (8), the character string position at which the cost function value is minimized is calculated.
However, in the following formula (8), I is the number of character strings, and i is an index number that identifies the character string.
Overlap _String (i) is a function indicating the overlap of character strings, Angle (i) is a function indicating the maximum angle of road links, and Dist (i) is a character string arrangement start candidate point and an input character string arrangement. It is a function indicating the distance from the start point. Α ₁ , α ₂ , and α ₃ are parameters for adjusting the respective function values. As described above, it is assumed that the parameters α ₁ , α ₂ , and α ₃ are all “1”.
Further, f (i) = α ₁ Overlap _String (i) + α ₂ Angle (i) + α ₃ Dist (i).
f = Σ [i = 0, I] f (i) (8)

FIG. 35 is a flowchart showing the flow of the calculation process of the character string arrangement that minimizes the cost function value, and shows the details of the process of step ST3 in FIG. Note that FIG. 15 is referred to for the configuration of the minimum cost function value calculation unit 4.
First, an index number that can take a value from 0 to the number of combinations of the character string arrangement start candidate points for each combination of the character string arrangement start candidate points determined to be arranged by the character string arrangement start candidate point creation unit 3 It is expressed by a combination of na (na).
The minimum cost function value calculation unit 4 sequentially performs processing from the combination (0) where na = 0 among all combinations of the character string arrangement start candidate points input from the character string arrangement start candidate point creation unit 3. (Step ST1e). The initial value of the minimum cost function value is set to MAX, which is the maximum value that can be handled by the computer functioning as the character string arrangement device 1.

In the specific example shown in FIG. 34, the character string arrangement start candidate points of character string 0 are Q0 (0, 20) and Q1 (8, 20) by the processing up to the above, and the character string arrangement start candidate of character string 1 As shown in FIG. 34, the points are Q5 (30, 0) and Q6 (30, 6).
Therefore, combinations of character string arrangement start candidate points are {Q0, Q5}, {Q0, Q6}, {Q1, Q5}, {Q1, Q6}.
The minimum cost function value calculation unit 4 first calculates a cost function value for the combination {Q0, Q5} (combination (0)).

The overlapping function calculation unit 21 between character strings performs overlapping function Over _String (i) between the character strings in the cost function f for all the character strings along the road link corresponding to the combination (na) of the character string arrangement start candidate points. Are respectively calculated (step ST2e).
Here, the duplication function calculation unit 21 between character strings calculates the degree of duplication between a character string for which the degree of duplication is first calculated and other character strings. When calculating the degree of duplication from the character string 0, duplication with the character string 1 that is a character string other than the character string 0 is searched.

  Next, the road link angle function calculation unit 22 indicates the angle of the road link along the character string in the cost function f for all the character strings along the road link corresponding to the combination (na) of the character string arrangement start candidate points. The value of the function Angle (i) is calculated (step ST3e). Here, the road link angle function calculation unit 22 calculates the angle formed by each link of the road links {A, B, C, D} along which the character string 0 is along.

  Next, the distance function calculation unit 23 includes each character string arrangement start candidate point of the combination (na) of character string arrangement start candidate points, and the character string arrangement start point acquired from the character string data by the character string data acquisition unit 2. The value of the function Dist (i) indicating the distance is calculated (step ST4e). Here, the distance between the character string arrangement start candidate points Q0 and Q1 and the character string arrangement start point for the character string 0, and the character string arrangement start candidate points Q5 and Q6 and the character string arrangement start for the character string 1 The distance to the point is calculated.

In step ST5e, the character string duplication function calculation unit 21, the road link angle function calculation unit 22, and the distance function calculation unit 23 perform all the characters along the road link corresponding to the combination (na) of the character string arrangement start candidate points. It is determined whether or not processing has been performed on the column (i).
If there is an unexecuted character string (i) (step ST5e; YES), the index number i is incremented by 1 (step ST7e), and the processing after step ST2e is repeated for the next character string.

If there is no unexecuted character string (i) (step ST5e; NO), the function value adding unit 24 uses the overlapping function calculating unit 21, the road link angle function calculating unit 22, and the distance function calculating unit 23 between the character strings. The calculated function values are added together to calculate a cost function value corresponding to the combination (na) of character string arrangement start candidate points (step ST6e).
Next, the function value adding unit 24 determines whether or not the cost function value calculated in step ST6e is the minimum (step ST8e).
Here, the cost function value determined last time and the cost function value determined this time are compared in magnitude, and if the cost function value determined this time is small (step ST8e; YES), the function value adding unit 24 has the smallest cost function. The value and the combination (na) corresponding to the cost function value are updated (step ST9e).

If the cost function value determined this time is large (step ST8e; NO), the function value adding unit 24 determines whether or not the cost function calculation processing has been performed for all combinations (na) of the character string arrangement start candidate points. Determination is made (step ST10e). If there is no unimplemented combination (na) (step ST10e; NO), the process is terminated.
If there is an unimplemented combination (na) (step ST10e; YES), +1 is added to the index number na (step ST11e), and the processes after step ST2e are repeated.

(2-1) Duplicate Function Value Calculation Processing Between Character Strings FIG. 36 is a flowchart showing the flow of calculation processing of overlapping function values between character strings, and shows details of the processing of step ST2e in FIG. Yes.
Each character string along the road link corresponding to the combination (na) of character string arrangement start candidate points is represented by a character string (i) with an index number i that can take a value from 0 to the number of character strings.
Further, among character strings along the road link corresponding to the combination (na) of the character string arrangement start candidate points, each character string other than the character string for which the degree of duplication is calculated is calculated from 0 for the degree of duplication. It is expressed by a character string (ia) having an index number ia that can take values up to the number of character strings other than the character string.
Furthermore, the duplication function calculation unit 21 between characters has a counter that counts the duplication number of character strings, and the value of the counter is represented by a parameter count.

First, the character overlap function calculation unit 21 performs subsequent processing in order from the character string (0) of i = 0 among the character strings along the road link corresponding to the combination (na) of character string arrangement start candidate points. Implement (step ST1f). At this time, count = 0 is set as the initial value of the counter.
Next, the character string upward arrangement unit 25a of the character overlap function calculation unit 21 arranges the character string (i) upward along the road link corresponding to the combination (na) of the character string arrangement start candidate points ( Step ST2f). Thereafter, the character string downward arrangement unit 25b arranges the character string (i) downward along the road link corresponding to the combination (na) of character string arrangement start candidate points (step ST3f). In this case, since the character string to be subjected to overlap calculation is the character string 0, the character string 0 is arranged upward and downward starting from the character string arrangement start candidate point Q0.
Note that the upward arrangement and the downward arrangement are performed by the same processing as that described with reference to FIG.

Similarly, the character string upward arrangement unit 25a arranges character strings (ia) other than the character string (i) upward along the road link corresponding to the combination (na) of character string arrangement start candidate points (step ST4f). The subsequent processing is performed in order from the character string (0) of ia = 0.
Next, the character string downward arrangement unit 25b arranges the character string (ia) other than the character string (i) downward along the road link corresponding to the combination (na) of character string arrangement start candidate points (step ST5f). . In this case, since the character string other than the character string to be subjected to the overlap calculation is the character string 1, the character string 1 is arranged upward and downward from the character string arrangement start candidate point Q5.

  Here, each character in the character string (i) is represented by a character (la) having an index number la that can take a value from 0 to the number of characters in the character string (i), and each character in the character string (ia) is represented by It is represented by a character (lb) having an index number lb that can take a value from 0 to the number of characters of the character string (ia). The character string overlap calculation unit 25c performs subsequent processing in order from the character (0) of la = 0 among the characters of the character string (i) and the character (0) of lb = 0 of the characters of the character string (ia). (Step ST6f).

Next, the character string duplication calculation unit 25c includes the character (la) of the character string (i) and the character (lb) of the character string (ia) that is one of the character strings other than the character string (i). Is determined (step ST7f). Here, the character string (i) is the character string 0, and the character string (ia) is the character string 1. FIG. 37 is a diagram showing a case where the character string 0 is arranged upward and downward with the character string arrangement start candidate point Q0 as the starting point, and the character string 1 is arranged upward and downward with the character string arrangement start candidate point Q5 as the starting point. is there. The character string overlap calculation unit 25c determines the overlap between the 0th character (la = 0) in the character string 0 and the 0th character (lb = 0) in the character string 1.
First, the duplication of the 0th (la = 0) upwardly arranged character “sentence” of the character string 0 and the 0th (lb = 0) upwardly arranged character “sentence” of the character string 1 is checked. .
Next, the duplication of the 0th (la = 0) upwardly arranged character “sentence” of the character string 0 and the 0th (lb = 0) downwardly arranged character “1” of the character string 1 is examined. It is done.
Similarly, the 0th (la = 0) downwardly arranged character “0” of the character string 0 and the 0th (lb = 0) upwardly arranged character “sentence” of the character string 1 are overlapped. Be examined.
The duplication of the 0th (la = 0) downwardly arranged character “0” of the character string 0 and the 0th (lb = 0) upwardly arranged character “sentence” of the character string 1 is checked.
If there is no duplication as a result of the duplication investigation four times in total, the character string duplication calculation unit 25c determines that there is no duplication, and proceeds to the process of step ST8f. On the other hand, if there is duplication, the character string duplication calculation unit 25c sets the determination result as duplication, and adds +1 to the count value count of the counter (step ST9f).
In this case, the 0th character (la = 0) in the character string 0 and the 0th character (lb = 0) in the character string 1 are determined not to overlap.

In step ST8f, the character string duplication calculation unit 25c determines whether or not duplication of all characters (lb) in the character string (ia) has been investigated. Here, if there is an unexamined character (lb) (step ST8f; YES), the character string overlap calculation unit 25c adds +1 to the index number lb (step ST10f), returns to step ST7f, and returns the character string 0. A duplication check between the 0th character (la = 0) and the next character (lb) is performed.
Here, the duplication of the 0th character (la = 0) in the character string 0 and the first character (lb = 1) in the character string 1 is examined. In this case, since the arrangement is as shown in FIGS. 32 and 37, it is determined that there is no overlap between the 0th character (la = 0) of the character string 0 and the first character (lb = 1) of the character string 1. Is done.

  When there is no unexamined character (lb) in the character string 1 (step ST8f; NO), the character string duplication calculation unit 25c determines whether or not duplication investigation has been performed for all characters (la) in the character string (i). (Step ST11f). If there is an unexamined character (la) (step ST11f; YES), the character string duplication calculation unit 25c adds +1 to the index number la (step ST12f), and returns to step ST7f to return the character string 0. A duplication check between the next character and the character (lb) in the character string 1 is performed. Here, the duplication of the first character (la = 1) of the character string 0 and the character (lb) of the character string 1 is examined.

  When there is no unexamined character (la) in the character string 0 (step ST11f; NO), the character string duplication calculation unit 25c determines whether all character strings (ia) other than the character string (i) have been investigated. Is determined (step ST13f). If there is an unexamined character string (ia) (step ST13f; YES), the character string overlap calculation unit 25c adds +1 to the index number ia (step ST14f), and returns to step ST4f to return the character string ( Conduct duplication survey with ia = 1). In this case, since there is no overlap between the character string 0 and the character string 1, the duplication number of the character string 0 as the counter value is “0”.

When there is no unexamined character string (ia) (step ST13f; NO), the character string duplication calculation unit 25c uses the duplication number of the character string (i) to be subjected to duplication calculation to calculate a character from the following equation (9). The value of the overlap function Over _String (i) between columns is calculated (step ST15f).
Note that the duplication number (i) in the following formula (9) is the duplication number of the character string (i). In this case, since the overlapping number of the character string 0 is “0”, Over _String (0) = 0. Similarly, Over _String (i) is calculated for other character strings.
if Duplicate (i) ≧ Overlap _String Threshold
Overlap _String (i) = 100
else
Overlap _String (i)
= Multiple (i) × 100 / Overlap _String Threshold
... (9)

(2-2) Road Link Angle Function Value Calculation Process FIG. 38 is a flowchart showing the flow of the road link angle function value calculation process, and shows the details of the process of step ST3e in FIG.
Each character string along the road link corresponding to the combination of character string arrangement start candidate points is represented by a character string (i) with an index number i that can take a value from 0 to the number of character strings.
Further, each node constituting the road link along which the character string (i) follows is specified by an index number lc that can take a value from 0 to the number of nodes of the road link−2.
Furthermore, the road link angle function calculation unit 22 represents the angle formed by the road link along the character string (i) as Angle (i).

First, the road link angle function calculation unit 22 performs the following processing in order from lc = 0 among the nodes (lc) in the road link along which the character string (i) follows (step ST1g). At this time, the angle Angle (i) = 0 is set.
Next, the road link angle function calculation unit 22 uses the three nodes in the road link along which the character string (i) follows and uses the following formula (10) to calculate the two links connected to each of the three adjacent nodes. Angle _lc (i) indicating the angle is calculated (step ST2g). However, in the following formula (10), n _lc , n _{lc + 1} , and n _{lc + 2} are three nodes that are continuously arranged in the road link along the character string (i). In the example of FIG. 20, the angle formed by the link {A, B} and the link {B, C} connecting the three nodes A, B, and C successively arranged in the road link along the character string 0 is π / 2. Therefore, Angle (0) = 50.
Angle _lc (i) = {π−∠ (m _lc , m _{lc + 1} , m _{lc + 2} )} × 100 / π
... (10)

Next, the road link angle function calculator 22 calculates the value of the function Angle _lc (i) calculated this time in the road link along the character string (i) as shown in the following formula (11). It is determined whether or not the function value indicating the angle formed by the two links connecting the three nodes to be calculated is larger than the set Angle (i) (step ST3g).
Here, if the value of Angle _lc (i) is larger than the value of Angle (i) (step ST3g; YES), the road link angle function calculation unit 22 calculates the value of the function Angle (i) as the function calculated this time. Update with the value of Angle _lc (i) (step ST4g).
In the case of the first time, Angle (i) is set to “0” in step ST1g, and therefore, i = 0 Angle (0) <Angle _A (0). Therefore, the value of Angle (i) is updated with the value of Angle _A (0).
Angle (i) = max {Angle _lc (i)} (11)

If the value of Angle _lc (i) is equal to or smaller than Angle (i) (step ST3g; NO), or the processing of step ST4g is completed, the road link angle function calculation unit 22 calculates the road along which the character string (i) follows. It is determined whether or not the above processing has been performed for all nodes in the link (step ST5g).
Here, when the index number lc is less than the number of nodes of road links along which the character string (i) follows −2, and the road link along which the character string (i) follows has a node that has not been subjected to the above processing (step ST5g; YES), the road link angle function calculation unit 22 adds +1 to the index number lc (step ST6g), and performs the processing after step ST2g for the next three nodes in the road link along which the character string (i) follows. To do.
In the example shown in FIG. 20, nodes B, C, and D are specified next to nodes A, B, and C, and the angle formed by the link {B, C} and the link {C, B} connecting them is expressed by the above formula. It is calculated using (10). In this case, since Angle _B (0) = 50, Angle (0) = Angle _B (0). Therefore, the value of Angle (0) is not updated.

As described above, Angle _lc (i) is repeated until it is determined that the above processing has been performed for all the nodes in the road link along which the character string (i) follows (Step ST5g; NO). The value of Angle (i) is updated according to In FIG. 20, the function Angle (0) indicating the angle at the road link along which the character string 0 follows is 50.

(2-3) Calculation processing of function value indicating distance between character string arrangement start candidate point and character string arrangement start point The distance between the character string arrangement start point and the character string arrangement start candidate point in step ST4e in FIG. Details of the calculation processing of the function shown will be described.
The distance function calculation unit 23 performs a character string arrangement start point (hereinafter referred to as an input character string arrangement start point) for the character string (i) preset in the character string data according to the following formula (12), A value of a function Dist (i) indicating a distance from the arrangement start candidate point is calculated.
However, (Q _xi , Q _yi ) in the following formula (12) is the position of the character string arrangement start candidate point of the character string (i). Further, (S _xi , S _yi ) is an input character string arrangement start point of the character string (i). In the case shown in FIG. 20, since the character string arrangement start candidate point of the character string 0 is Q0, (Q _x0 , Q _y0 ) = Q0 (0, 20). As described above, the character string arrangement start point I of the character string 0 is (S _x0 , S _y0 ) = I (12, 10).
Therefore, Dist (0) = √244 (square root of 244) from the following equation (12).

After performing the above-described processing also on the character string 1, the function value adding unit 24 calculates the cost function value f according to the above equation (8). In the character string 0 and the character string 1, the following relationship is established.
f (0) = Overlap _String (0) + Angle (0) + Dist (0)
= 50 + √244
f (1) = Overlapping _String (1) + Angle (1) + Dist (1)
= 12
Therefore, the cost function value f is f = f (0) + f (1) = 62 + √244.

  The above-described processing is performed for other combinations {Q0, Q6}, {Q1, Q5}, {Q1, Q6} of character string arrangement start candidate points. As a result, the cost function value is as shown in FIG. In this case, the minimum cost function value calculation means 4 determines that the cost function value f is minimum when {Q0, Q6}. Therefore, {Q0, Q6} is stored in the output character string arrangement start candidate point storage area 5.

  When the character string arrangement device 1 according to the first embodiment arranges a character string on the screen, the character string arrangement start candidate point set stored in the output character string arrangement start candidate point storage unit 5 is used to Character strings can be arranged at positions where there is little overlap between columns. In the above-described example, the character string arrangement start candidate point set {Q0, Q6} is used for character string arrangement.

As described above, according to the first embodiment, the character string data acquisition unit 2 that acquires data of the character string to be arranged on the display screen, and the arrangement of the character strings acquired by the character string data acquisition unit 2 A character string arrangement start candidate point creation unit 3 for creating a character string arrangement start candidate point on the display screen to be started, and a character string arrangement start candidate point created by the character string arrangement start candidate point creation unit 3 as a starting point Factors that affect the readability of the character strings based on the positional relationship between the character strings in each of the arrangements in such a way that the readings are in the forward direction in the opposite directions above. A minimum cost function value calculating unit 4 for calculating a cost function value for evaluating an overlap, an angle formed by a road link, and a distance between an input character string arrangement start point and a character string arrangement start candidate point, and Position candidate And an output string arrangement candidate start point storage unit 5 for storing a string arrangement candidate start point for changing the starting position.
By configuring in this way, for example, by minimizing the cost function value including the function indicating the duplication of the character strings arranged upward and downward with respect to the road, the position where the duplication of the character strings is small in any map orientation. Strings can be placed.
As a result, even if the orientation of the map changes due to a change in the position of the vehicle as in a car navigation system, it is possible to suppress the display of overlapping character strings on the display screen, and the character strings can be easily Can be recognized.

Embodiment 2. FIG.
In the first embodiment, by minimizing the cost function value indicating the duplication of the character string pairs arranged upward and downward with the road link in between, the character strings overlap each other regardless of the map orientation. An apparatus has been shown in which character string arrangement start candidate points that are few positions are determined, and character strings are rearranged based on the character string arrangement start candidate points.
In the second embodiment, a mode will be described in which not only overlapping character strings but also a position where there is little overlap between character strings and symbols is determined as a character string arrangement start candidate point. The second embodiment also differs from the first embodiment in that a genetic algorithm (hereinafter referred to as GA) is used when obtaining the minimum value of the cost function f.

FIG. 40 is a block diagram showing a configuration of a character string arrangement device according to Embodiment 2 of the present invention. As in the first embodiment, the character string arranging device 1A according to the second embodiment is a device that constitutes a map display device used for, for example, a navigation device, and is a link that represents a road on a map. A character string indicating information such as the name is arranged along the line. Also in the following description, a case where the character string arranging device 1A arranges a character string indicating a road name or the like along a road link of the road will be described.
In FIG. 40, a character string arrangement device 1A according to Embodiment 2 includes a character string data acquisition unit 2, a character string arrangement start candidate point creation unit 3, an output character string arrangement start candidate point storage unit 5a, a symbol data acquisition unit 26, An individual initialization unit 27, a cost function value calculation unit 28 for each individual, a cost function value minimum individual storage unit 29, a convergence determination unit 30, an individual re-creation unit 31, a crossover unit 32, and a mutation unit 33 are provided. The constituent units 27 to 33 extract the minimum cost function value using the GA.

The character string data acquisition unit 2 is a component that acquires character string data from an external storage device. The character string data is, for example, a character string of a road name to be displayed on a map screen, road link data indicating a road link along which the character string is included, and a character string on the map screen. Is a data including a character string arrangement start point indicating a position where the arrangement of the character string starts.
The symbol data acquisition unit 26 is a component that acquires symbol data from an external storage device. The symbol data is, for example, data indicating a symbol included in map data used for map display and displayed on a road or along a road on a map screen, and a symbol arrangement indicating a position where the symbol is arranged Data including points. In the case of a navigation device, a symbol indicating the current position of the host vehicle can be mentioned.

The character string arrangement start candidate point creation unit 3 starts arrangement of the character string on the map screen when the character string arrangement start point of the character string acquired from the character string data by the character string data acquisition unit 2 is changed. This is a component that creates point candidates (hereinafter referred to as character string arrangement start candidate points).
The individual initialization unit 27 creates a combination of character string placement start candidate points to be processed first from a combination of character string placement start candidate points created by the character string placement start candidate point creation unit 3 and determined to be placeable. It is a component. In this way, the character string arrangement start candidate point is selected with equal probability from the character string arrangement start candidate point created by the character string arrangement start candidate point creation unit 3 for each character string, and the individual to be processed first is determined. Processing is called individual initialization. Details of this processing will be described later with reference to FIG.

  The cost function value calculation unit 28 for each individual has a factor that affects the readability of the character string in a pair of a character string arranged upward from a character string arrangement start candidate point for each individual and a character string arranged downward. A function that indicates the overlap between character strings in each layout, a function that indicates the angle between adjacent road links, a function that indicates the distance between the character string layout start point and the text string layout start candidate point obtained from the string data, and a symbol The cost function value is calculated by adding together the function values indicating the overlap between the character string and the character string, and the individual having the smallest cost function value and the cost function value are obtained.

  The cost function value minimum individual storage unit 29 is a component that stores the minimum value of the cost function value calculated by the cost function value calculation unit 28 for each individual and the individual, and stores the cost function value for each stored individual. Among the minimum values, the minimum cost function and its individual are stored in the output character string arrangement start candidate point storage unit 5a. The output character string arrangement start candidate point storage unit 5a is a storage unit that stores the minimum cost function value and its individual among the individual stored in the cost function value minimum individual storage unit 29 and its cost function value.

The convergence determination unit 30 is a configuration unit that determines whether or not the minimum value of the cost function value for each individual stored in the cost function value minimum individual storage unit 29 has converged uniformly.
The individual recreating unit 31 selects a predetermined number of individuals by selecting a character string arrangement start candidate point with a predetermined selection probability from the individual initialized by the individual initializing unit 27 or the individual previously created by itself. This is a component to be recreated.
The crossover unit 32 further selects two individuals from a plurality of individuals to be crossed selected with a predetermined probability (crossover probability CROSSOVER) from the individuals created by the individual recreating unit 31, and a character string between the selected individuals. This is a component that performs a so-called crossover process in which a part of the combination of arrangement start candidate points is replaced.
The mutation unit 33 sets the character string arrangement start candidate points of the individual created by the individual re-creation unit 31 and the individual subjected to the crossover process by the crossover unit 32 with a predetermined probability (mutation probability Mutation). This is a configuration unit for converting to another character string arrangement start candidate point selected from the character string arrangement start candidate points registered in the individual.

  The character string data acquisition unit 2, the character string arrangement start candidate point creation unit 3, the output character string arrangement start candidate point storage unit 5a, the symbol data acquisition unit 26, the individual initialization unit 27, and the cost function value calculation unit for each individual 28, the cost function value minimum individual storage unit 29, the convergence determination unit 30, the individual re-creation unit 31, the crossover unit 32, and the mutation unit 33, for example, cause a computer to execute a character string arrangement program according to the gist of the present invention. This is realized as a specific means in which hardware and software cooperate. That is, each function of the components 2, 3, 26 to 33 is defined in each program module constituting the character string arrangement program. Moreover, the output character string arrangement | positioning start candidate point memory | storage part 5a is constructed | assembled on the storage area of the memory | storage device mounted in said computer, for example. Examples of the storage device include a hard disk device, a semiconductor memory, a drive device for storage media such as a CD and a DVD, and the like.

41 is a block diagram showing the configuration of the cost function value calculation unit for each individual in FIG. 40, and FIG. 41 (a) shows the overall configuration of the cost function value calculation unit 28 for each individual. 41 (b) shows the configuration of the symbol and character string overlap function calculator 34 in FIG. 41 (a). In FIG. 41A, the cost function value calculation unit 28 for each individual includes a duplication function calculation unit 21a between character strings, a road link angle function calculation unit 22a, a distance function calculation unit 23a, and a symbol and character string duplication function calculation. Unit 34 and a function value adding unit 24a.
As shown in FIG. 41 (b), the symbol / character string overlap function calculator 34 includes a circumscribed rectangle creating unit 35, a circumscribed rectangle / symbol overlap determining unit 36, and a circumscribed rectangle / symbol overlap function calculating unit. 37.

The duplication function calculation unit 21a between character strings is a character string when the character string placement start candidate point of the individual initialized by the individual initialization unit 27 or the individual recreated unit 31 is placed as a starting point. This is a configuration unit that determines the overlap between characters using a function that indicates the overlap between the character strings.
The road link angle function calculation unit 22a is a component that calculates the value of the function Angle (i) that indicates the angle formed by the road link along which the character string acquired from the character string data follows.
The distance function calculation unit 23 a is created by the character string arrangement start point of the character string acquired from the character string data by the character string data acquisition unit 2 and the individual or individual re-creation unit 31 initialized by the individual initialization unit 27. This is a component that calculates the value of the function Dist (i) indicating the distance from the character string arrangement start candidate point of the individual.
Function value plus combined unit 24a functions _{Overlap String} (i), the value of the function Angle (i), the function Dist (i) and function _{Overlap symbol} (i) the sum total of consisting cost function f (i), the individual initial The character string arranged at the character string arrangement start candidate point of the individual initialized by the conversion unit 27 or the individual created by the individual re-creation unit 31 is calculated, and the value of the cost function f (i) is minimized. It is a component that determines an individual and its cost function value. The individual of the minimum cost function value determined by the function value adding unit 24 a and its cost function value are stored in the cost function value minimum individual storage unit 29.

The symbol and character string duplication function calculation unit 34 uses the character string arrangement start candidate points of the individual initialized by the individual initialization unit 27 or the individual re-creation unit 31 to arrange characters as the starting points. This is a configuration unit that determines an overlap between a string and a symbol using a function indicating an overlap between a character string and a symbol.
In addition, the circumscribed rectangle creating unit 35 arranges characters constituting the character string upward and downward by two characters via a road link along which the character string to be processed follows, and these four characters are arranged upward and downward. This is a configuration unit that creates a circumscribed rectangle that circumscribes.
The circumscribed rectangle and symbol overlap determining unit 36 is a component that determines the overlap between each symbol acquired from the symbol data by the symbol data acquiring unit 26 and the circumscribed rectangle created by the circumscribed rectangle creating unit 35.
The circumscribed rectangle and symbol overlap function calculator 37 uses the circumscribed rectangle and symbol overlap function obtained by the circumscribed rectangle and symbol overlap determining unit 36 to determine the overlap function of the circumscribed rectangle and symbol Overlap _symbol (i). ) For calculating the value of.

42 is a block diagram showing the configuration of the individual re-creation unit in FIG. 42, the individual recreating unit 31 includes an individual ordering unit 38 and a recreating unit 39.
The individual ordering unit 38 performs the order according to the cost function value calculated by the cost function value calculation unit 28 for each individual, with respect to the individual initialized by the individual initialization unit 27 or the individual previously created by itself. It is the component which provides.
The recreating unit 39 uses the individual selected by the body initializing unit 27 or the previous time with a predetermined selection probability determined using the order given by the individual ordering unit 38 and the cost function value corresponding thereto. The configuration unit re-creates a predetermined number of individuals by selecting a character string arrangement start candidate point from the created individuals. The method for calculating the selection probability will be described later with reference to FIG.

FIG. 43 is a block diagram showing the configuration of the crossing part in FIG. 43, the crossover unit 32 includes a crossover target individual determination unit 40, a two individual determination unit 41, a crossover position determination unit 42, and a two-individual crossover unit 43.
The crossover target individual determination unit 40 is a configuration unit that selects a crossover target individual that is a target for performing crossover processing with a predetermined probability (crossover probability CROSSOVER) from the individuals created by the individual recreating unit 31.
The two-individual determination unit 41 is a configuration unit that determines two individuals to perform crossover processing from the cross-target individual selected by the cross-target individual determination unit 40.
The crossover position determination unit 42 is a component that determines a character string arrangement start candidate point (crossover position in the individual) for executing the crossover process among the two individuals determined by the two-individual determination unit 41.
The two-individual crossover unit 43 is a component that executes a crossover process of replacing the character string arrangement start candidate point that is the crossover position determined by the crossover position determination unit 42 between the two individuals determined by the two-individual determination unit 41. is there.

FIG. 44 is a diagram illustrating an example of an individual handled in the second embodiment. As shown in FIG. 44, an individual is a combination of character string arrangement start candidate points that are created for each character string by the character string arrangement start candidate point creation unit 3 and determined to be arranged. In the example of FIG. 44, the individual 0 combines Q0 of the character string arrangement start candidate points Q0 and Q1 of the character string 0 and Q6 of the character string arrangement start candidate points Q5 and Q6 of the character string 1. Composed.
Further, as shown in FIG. 44, the individual is data in which the character string arrangement start candidate points are arranged in the order close to the origin. That is, since the character string arrangement start candidate point Q0 of the character string 0 is closer to the origin than the character string arrangement start candidate point Q6 of the character string 1, the individual 0 is data in which the arrangement of Q0 and Q6 is defined.

Next, the operation will be described.
FIG. 45 is a flowchart showing a flow of operations performed by the character string arrangement device according to the second embodiment. The outline of the process of acquiring the character string arrangement start candidate point will be described with reference to FIG.
First, the character string data acquisition unit 2 acquires character string data from an external storage device such as a map database, and all character strings included in the character string data, and road links and character strings corresponding to the respective character strings. An arrangement start point is extracted (step ST1h).
Next, the symbol data acquisition unit 26 acquires symbol data from an external storage device such as a map database, and extracts all symbols included in the symbol data and their arrangement positions (step ST2h).

In the following, it is assumed that the character string data acquisition unit 2 has acquired two character strings of character string 0 and character string 1 from the character string data, and character string 0 and character string 1 and road links corresponding to these character strings. The character string arrangement start point is assumed to be defined as follows.
FIG. 46 is a diagram showing an example of the relationship between the road link along the character string 0 and the character string 1, the character string arrangement start point, and the symbol position, and in the map screen in which the map orientation is north (north up) Is a two-dimensional coordinate system with x-axis and north-south direction as y-axis.
The road links along which the character string 0 follows are nodes A (0, 20), B (10, 20), C (10, 10), and D (30, 10). Link). In addition, it is assumed that the character string arrangement start point of the character string 0 preset in the character string data is I (12, 10).
In addition, the road links along the character string 1 connect the nodes E (30, 0), D (30, 10), and F (30, 30) with links (links shown by solid lines in FIG. 46). Consists of. It is assumed that the character string arrangement start point of the character string 1 preset in the character string data is J (30, 12).

Each of the characters “sentence”, “character”, “column”, and “0” included in the character string 0 has a vertical width of 5 and a horizontal width of 3, and the “sentence”, “ Similarly, each of the characters “character”, “column”, and “1” is assumed to have a vertical width of 5 and a horizontal width of 3.
It is assumed that the distance RCgap between the road link and each character of the character string 0 and the character string 1 is “3”, and the character-to-character distance CCgap is “√2 (square root of 2)”.

Furthermore, the character string arrangement start candidate point creation unit 3 creates five character string arrangement start candidate points PATTERN_NUM for each character string, and the character arrangement candidate point interval DIF_CAND_POINT for arranging each character of the character string. the "2" is assumed to be, the threshold _overlap string threshold of duplicate numbers between string is assumed to be "5".

Parameters α ₁ , α ₂ , α ₃ , and α ₄ when calculating cost function values to be described later are all set to “1”. The maximum values such as the above-mentioned intervals RCgap, CCgap, and DIF_CAND_POINT that can be processed by the computer functioning as the character string arranging device 1A are “MAX”.

Assume that two symbols, symbol 0 and symbol 1, have been acquired from the symbol data by the symbol data acquisition unit 26, the arrangement position of symbol 0 is K (5, 20) in FIG. 46, and the arrangement position of symbol 1 is Let L (30, 24).
Further, the number of individuals POPULATION_SIZE created by the individual initialization unit 27 and the individual re-creation unit 31 is set to “3”, the crossover probability CROSSOVER selected by the crossover unit 32 as a crossover target individual is set to 60%, and the mutation unit 33. As a mutation process, the mutation probability MUTATION for selecting an individual to convert the character string arrangement start candidate point is increased by 5%.
Further, CONVERGENCE_CONTINUOUS used for convergence determination by the convergence determination unit 30 is set to “5”, and CONVERGENCE_RATE is set to 50%.
Note that the above-described parameters and probability values are set in advance in the character string data for each character string.

  The character string arrangement start candidate point creation unit 3 creates character string arrangement start candidate points on the road links along which the respective character strings acquired by the character string data acquisition unit 2 follow (step ST3h). At this time, the character string arrangement start candidate point creation unit 3 determines whether or not the character string arrangement start candidate point can be arranged for each character of each character string. The method of creating the character string arrangement start candidate point and determining whether or not to arrange the character string arrangement start candidate point is the same as in the first embodiment. In this case, as in the first embodiment, character string arrangement start candidate points Q0 and Q1 for character string 0 and character string arrangement start candidate points Q5 and Q6 for character string 1 are created.

  Next, the individual initialization unit 27 combines POPULATION_SIZE character string arrangement start candidate points from the character string arrangement start candidate points created by the character string arrangement start candidate point creation unit 3 and determined to be arranged (individual ) Is created, so-called individual initialization is performed (step ST4h). Here, from the character string arrangement start candidate points created by the character string arrangement start candidate point creation unit 3, a character string arrangement start candidate point is selected with equal probability for each character string, and an individual to be processed first is determined. . Under the conditions assumed as the premise described above, Q0 and Q1 are created as character string arrangement start candidate points for character string 0, and Q5 and Q6 are created as character string arrangement start candidate points for character string 1. When the character string arrangement start candidate points are selected with equal probability every time, individuals 0 to 2 shown in FIG. 47 are created.

The cost function value calculation unit 28 for each individual overlaps or adjoins in pairs of a character string arranged upward and a character string arranged downward, starting from a character string arrangement start candidate point for each individual, according to the following equation (13). Deformation according to the angle between road links, the distance between the character string arrangement start point acquired from the character string data by the character string data acquisition unit 2 and the character string arrangement start candidate point, and the overlap between the symbol and the character string, respectively. A cost function including each function is calculated, and an individual having the minimum cost function value and its cost function value are calculated (step ST5h). The cost function value minimum individual storage unit 29 stores the minimum cost function value calculated by the cost function value calculation unit 28 for each individual and the individual (step ST6h).
Note that the parameter i in the following formula (13) indicates the index number of the character string, and Over _String (i), Angle (i), Dist (i), α ₁ , α ₂ , α ₃ are all It is assumed that the functions and variables are the same as those in the above equation (8).
Overlap _String (i) is a function indicating the overlap of character strings and symbols, and α ₄ is a parameter for adjusting the value of the function Over _symbol (i).
Where f ′ (i) = α ₁ Overlap _String (i) + α ₂ Angle (i) + α ₃ Dist (i) + α ₄ Overlap _symbol (i).
f ′ = Σ [i] f ′ (i) (13)

Next, the convergence determination unit 30 determines whether or not the minimum value of the cost function value for each individual stored in the cost function value minimum individual storage unit 29 has converged to a certain level (step ST7h).
As a method of convergence determination, when the individual stored in the cost function value minimum individual storage unit 29 does not change only CONVERGENENCE_CONTINUOUS times, or CONVERGENCE_RATE% or more in the individual stored in the cost function value minimum individual storage unit 29 If the individuals have the same cost function value, it is determined that convergence has occurred at the minimum value of the cost function value. In other cases, it is determined that there is no convergence.
If it is determined that there is convergence (step ST7h; YES), the process ends.
If it is determined that there is no convergence (step ST7h; NO), the process proceeds to step ST8h.

In step ST8h, the individual recreating unit 31 selects a character string arrangement start candidate point with a predetermined selection probability from the individual initialized by the individual initializing unit 27 or the individual previously created by itself. Recreate a number of individuals. In the individual re-creation method, first, the selection probability of an individual is calculated from the following equation (17).
In the following formula (17), l is an index number for specifying each individual, and g _l is the relative size of the cost function values of the individuals ordered by the individual ordering unit 38. Selection_prob _l indicates the probability that the individual l is selected as a new individual.

Next, the crossover unit 32 further selects two individuals from a plurality of individuals to be crossed selected from the individuals created by the individual recreating unit 31 with a predetermined crossover probability CROSSOVER, and a character string between the selected individuals. Crossover processing for exchanging a part of the combination of placement start candidate points is executed (step ST9h).
Next, the mutation unit 33 assigns the character string arrangement start candidate points of the individual created by the individual recreating unit 31 and the individual subjected to the crossover process by the crossover unit 32 to these individuals with a predetermined mutation probability Mutation. Conversion is made to another character string arrangement start candidate point selected from the registered character string arrangement start candidate points (step ST10h). When there are a plurality of other character string arrangement start candidate points, one of them is selected with equal probability.

Thereafter, the cost function value calculation unit 28 for each individual determines whether or not there is no convergence by the convergence determination unit 30, and determines whether or not the above-described processing has been executed a predetermined number of times (step ST11h). . Here, although it is determined by the convergence determination unit 30 that there is no convergence, when it is determined that the above-described processing has been executed a predetermined number of times (step ST11h; YES), the processing ends.
Further, when it is determined that the convergence determination unit 30 has no convergence and the number of times the above processing has been executed is less than the predetermined number of iterations (step ST11h; NO), the cost function value for each individual The calculating unit 28 inputs the individual created by the individual re-creating unit 31, the individual subjected to the crossover process by the crossover unit 32, and the individual subjected to the process by the mutation unit 33, and after step ST5h Repeat the process.

(1A) Cost Function Value Calculation Processing for Each Individual FIG. 48 is a flowchart showing the flow of cost function value calculation processing for each individual, and shows details of processing in step ST5h in FIG.
Here, each individual for which a cost function value is calculated is represented by an individual (nb) having an index number nb that can take a value from 0 to the number of individuals.
Also, each character string along the road link corresponding to the individual (nb) is represented by a character string (i) with an index number i that can take a value from 0 to the number of character strings.

The character overlap function calculation unit 21a performs the subsequent processing in order from the character string (0) of i = 0 among the character strings (i) corresponding to the individual (0) of nb = 0 (step ST1i). .
First, the character overlap function calculation unit 21a calculates the value of the function Over _String (i) indicating the overlap between character strings for the character string (i) corresponding to the individual (nb) (step ST2i). The method of calculating the value of the function Over _String (i) is the same as that of the character overlap function calculation unit 21 in the first embodiment.
In the individual 0, the individual 1, and the individual 2 shown in FIG. 47, as shown in the column (a) of FIG. 49, the character 0 and the character string 1 are 0 in the individual 0, and the character string 0 and the character string 1 are in the individual 1. 40 in the individual 2 and 40 in the character string 0 and the character string 1 in the individual 2.

Next, the road link angle function calculation unit 22a calculates the value of the function Angle (i) indicating the angle of the road link along the character string (i) for the character string (i) corresponding to the individual (nb) ( Step ST3i). The calculation method of the value of the function Angle (i) is the same as that of the road link angle function calculation unit 22 of the first embodiment.
In the individual 0, individual 1, and individual 2 shown in FIG. 47, as shown in the column (b) of FIG. 49, the character 0 is 50 for the individual 0, 0 is the character string 1, and the character string 0 is also the individual 1. 50, 0 for character string 1, 50 for character string 0 for individual 2, and 0 for character string 1.

Next, the distance function calculation unit 23a calculates the character string arrangement start candidate point of each individual (nb) and the character string arrangement start point of the character string (i) acquired from the character string data by the character string data acquisition unit 2. The value of the function Dist (i) indicating the distance is calculated (step ST4i). The calculation method of the value of the function Dist (i) is the same as that of the distance function calculation unit 23 of the first embodiment.
In the individual 0, the individual 1, and the individual 2 shown in FIG. 47, as shown in the column (c) of FIG. 49, the character string 0 in the individual 0 is √244, the character string 1 is 12, and the individual 1 is the character string 0. √116, 6 for character string 1, √116 for character string 0 in individual 2, and 12 for character string 1.

Thereafter, the symbol and character string overlap function calculation unit 34 calculates a function value indicating the overlap between the character string (i) corresponding to the individual (nb) and the symbol (step ST5i). Details of this processing will be described later with reference to FIG.
In step ST6i, the duplication function calculation unit 21a for character strings, the road link angle function calculation unit 22a, the distance function calculation unit 23a, and the duplication function calculation unit 34 for symbols and character strings include all characters corresponding to the individual (nb). It is determined whether or not processing has been performed on the column (i).
If there is an unexecuted character string (i) (step ST6i; YES), the index number i is incremented by +1 (step ST7i), and the processing after step ST2i is repeated for the next character string.

  If there is no unexecuted character string (i) (step ST6i; NO), the character string overlap function calculation unit 21a, the road link angle function calculation unit 22a, the distance function calculation unit 23a, and the symbol and character string overlap function calculation The unit 34 determines whether or not all the individuals (nb) have been processed (step ST8i). If there is an unexecuted individual (nb) (step ST8i; YES), the index number nb is incremented by +1 (step ST9i), and the processing after step ST2i is repeated for the next individual. If there is no unexecuted individual (nb) (step ST8i; NO), the process is terminated.

(1A-1) Symbol and Character String Duplicate Function Value Calculation Processing FIG. 50 is a flowchart showing the flow of symbol and character string overlap function value calculation processing. The details of the processing of step ST5i in FIG. Show.
Each character string corresponding to the individual (nb) is represented by a character string (i) having an index number i that can take a value from 0 to the number of character strings.
Further, a circumscribed rectangle consisting of two characters in each of the upwardly arranged and downwardly arranged character strings (i) is represented by a circumscribed rectangle (nc) having an index number nc that can take a value from 0 to the number of characters −1 of the character string (i). It expresses with.
Further, each symbol for which the degree of overlap is calculated is represented by a symbol (ka) having an index number ka that can take a value from 0 to the number of symbols.
Furthermore, the symbol and character string overlap function calculator 34 has a counter that counts the number of symbols that overlap the character string, and the value of the counter is represented by a parameter count.

First, the symbol and character string overlap function calculation unit 34 performs the following processing in order from the character string (0) of i = 0 among the character strings corresponding to the individual (0) of nb = 0 (step ST1j). ). At this time, count = 0 is set as the initial value of the counter, and nc = 0 is set.
Next, the circumscribed rectangle creation unit 31 arranges the character string (0) upward along the road link corresponding to the individual (0) (step ST2j). Subsequently, the circumscribed rectangle creating unit 31 arranges the character string (0) downward along the road link (step ST3j). Thereby, in Embodiment 1 described above, the arrangement as shown in FIG. 32 is obtained.

  Next, the circumscribed rectangle creating unit 31 specifies the left two characters in the character string (i) arranged upward and downward, and creates a circumscribed rectangle (nc) circumscribing these four characters (step ST4j). . In the case of the arrangement shown in FIG. 32, as shown in FIG. 51, the vertexes are T0 (0, 12) and T1 (0, 28) from each of the first two characters of the character string 0 arranged upward and downward. ), Circumscribed rectangle 0 (nc = 0) which is T2 (9, 28) and T3 (9, 12) is created. In FIG. 51, the circumscribed rectangle A3 is a circumscribed rectangle of each character of the character string 1 arranged upward and downward with the road link b as a boundary, and the circumscribed rectangle B3 is disposed upward and downward with the road link a as a boundary. It is a circumscribed rectangle of each character of the character string 0. 51, the symbol at the K position is symbol 0, and the symbol at the L position is symbol 1. As described in the description of step ST2h in FIG. 45, it is assumed that symbol 0 is located at K (5, 20) and symbol 1 is located at L (30, 24).

Thereafter, the circumscribed rectangle and symbol overlap determining unit 36 determines whether or not there is an overlap between the circumscribed rectangle (nc) and the symbol (ka) (step ST5j). Here, first, the circumscribed rectangle and symbol overlap determination unit 36 calculates overlap1 to overlap4 represented by the following equation (14). In the following formula (14), (x _rect ^TopLeft , y _rect ^TopLeft ) is the upper left coordinate of the circumscribed rectangle (nc), and (x _rect ^BottomLeft , y _rect ^BottomLeft ) is the lower left coordinate of the circumscribed rectangle (nc), x ^{_rect BottomRight,} _{y rect} ^BottomRight) is the lower right coordinates of the circumscribed rectangle ^{_{(nc), (x rect TopRight}} , y rect TopRight) shows the upper-right coordinates of the circumscribed rectangle (nc). Also, (x _symbol , y _symbol ) indicates the coordinates of the symbol.

Subsequently, the circumscribed rectangle and symbol overlap determination unit 36 determines that there is an overlap when the above overlap1 to overlap4 satisfy the following expression (15), and determines that there is no overlap otherwise. In the case of the example shown in FIG. 51, overlap1 = 80, overlap2 = 72, overlap3 = 64, overlap4 = 72, which satisfies the following expression (15). Therefore, it is determined that the circumscribed rectangle 0 (nc = 0) and the symbol 0 (ka = 0, symbol (0)) overlap.
overlap1> 0 &overlap2> 0 &overlap3> 0 &overlap4> 0 (15)

  If it is determined that there is no overlap between the circumscribed rectangle (nc) and the symbol (ka), the circumscribed rectangle and symbol overlap determining unit 36 performs the above processing for all circumscribed rectangles (nc) that can be created from the character string (i). Is determined (step ST6j). Here, when the index number nc does not reach the number of characters −1 in the character string (i) and it is determined that there is an unprocessed circumscribed rectangle (nc) (step ST6j; YES), the index number nc is incremented by +1. Then (step ST7j), the process returns to step ST4j, and the next circumscribed rectangle and symbol overlap are determined.

On the other hand, if it is determined in step ST5j that there is an overlap between the circumscribed rectangle (nc) and the symbol (ka), the circumscribed rectangle and symbol overlap determining unit 36 counts the number of symbols that overlap with the character string (i). The value count is incremented by +1 (step ST8j).
When the index number nc reaches the number of characters −1 of the character string (i) and it is determined that there is no unprocessed circumscribed rectangle (nc) (step ST6j; NO), or when the process of step ST8j is completed, the circumscribed rectangle and symbol The duplication determination unit 36 determines whether or not the above processing has been executed for all symbols (ka) (step ST9j). If it is determined that the index number ka reaches the number of symbols and there is no unprocessed symbol (ka) (step ST9j; NO), the process proceeds to step ST10j.

  When it is determined that the index number ka does not reach the number of symbols and there is an unprocessed symbol (ka) (step ST9j; YES), the index number ka is incremented by 1 (step ST11j), and the process returns to step ST4j. The overlap between the next symbol and the character string (i) is determined. At this time, in the example shown in FIG. 51, the overlap between the character string 0 and the symbol 1 (ka = 1, symbol (1)) is determined. First, circumscribed rectangle 0 is created for character string 0 in the same manner as described above, and the overlap between circumscribed rectangle 0 and symbol 1 is determined. In this case, since the coordinates of the symbol 1 are L (30, 24), overlap1 = 480, overlap2 = 36, overlap3 = −336, overlap4 = 108, and the above equation (15) is not satisfied. For this reason, it is determined that circumscribed rectangle 0 and symbol 1 do not overlap.

The number of symbols overlapping with the character string 0 is determined by repeating the above process. In the case of FIG. 51, symbol 0 is the only symbol overlapping with character string 0. The number of symbols (ka) overlapping with the character string (i) is the overlapping number of character strings and symbols. Therefore, the overlapping number of the character string 0 and the symbol is “1”.
Thereafter, the circumscribed rectangle / symbol overlap function calculation unit 37 calculates the circumscribed rectangle / symbol overlap function using the following equation (16) (step ST10j). In addition, the overlap _symbol number (i) in the following formula (16) indicates the overlapping number of the character string i and the symbol. In the example of FIG. 51, the number of overlap _symbols (0) is 1, and the overlap _symbol (0) is 20.

if overlap _symbol number (i) ≧ Overlap _symbol Threshold
overlap _symbol (i) = 100
else
overlap _symbol (i)
= Overlap _symbol number (i) × 100 / (Overlap _symbol Threshold)
... (16)

In the same manner as described above, a duplication function of a character string and a symbol is calculated for the next character string and individual. As a result, as shown in the column (d) of FIG. 49, the overlap _symbol number of the character string 0 and the symbol of the character string 1 is 20 for the character string 0 for the individual 0, 0 for the character string 1, and the character number for the individual 1 0 in column 0, 20 in character string 1, 0 in character string 0 in individual 2, and 0 in character string 1.

Thereafter, the function value adding unit 24a adds the cost functions according to the above equation (13). In this case, the value of the cost function f of each individual is 82 + √244 for the individual 0, 156 + √116 for the individual 1, and 142 + √116 for the individual 2 as shown in the column (e) of FIG.
The cost function value minimum individual storage unit 29 holds each individual 0 to 2 and the value of the cost function f. Among these, the individual having the lowest value of the cost function f and the cost function value are output as output characters. Store in the column arrangement start candidate point storage unit 5a. In the above example, since the value of the cost function f of the individual 0 is the lowest, the character string arrangement start candidate points Q0 and Q5 of the individual 0 and the cost function value 82 + √244 of the individual 0 are output character string arrangement start candidate points. It is stored in the storage unit 5a.

(2A) Individual Re-creation Process Here, the individual re-creation process in step ST8h in FIG. 45 will be described in detail.
First, the individual ordering unit 38 orders the individual stored in the cost function value minimum individual storage unit 29 in descending order of the value of the cost function f.
In the above example, the value of the cost function f of the individual 0 is 82 + √244, the value of the cost function f of the individual 1 is 156 + √116, and the value of the cost function f of the individual 2 is 142 + √116. Is 3, individual 1 is 2, and individual 3 is 1.

Next, the individual re-creating unit 39 re-creates POPULATION_SIZE individuals according to the above equation (17). In the above example, since the order of the individual 0 is 3, the order of the individual 1 is 2, and the order of the individual 3 is 1, the selection probability Selection_prob is 0 for the individual 0 as shown in the column (f) of FIG. 3/6, 1/6 for individual 1 and 2/6 for individual 2.
In the following description, by creating a new individual with the selection probability shown in the column (f) of FIG. 49 from the individual created by the individual initialization unit 27, for example, an individual as shown in FIG. It shall be created. In addition, since a new individual is created by selecting a character string arrangement start candidate point from an old individual stochastically, an individual as shown in FIG. 52 is not necessarily created every time.

(3A) Crossover Process FIG. 53 is a flowchart showing the flow of the crossover process, and shows details of the process of step ST9h in FIG. FIG. 43 is referred to for the configuration of the crossing section 32. In addition, the number of crossover target individuals in the individual created by the individual initialization unit 27 is expressed by an index number nd that can take a value from 0 to the number of crossover target individuals.
First, the crossover target individual determination unit 40 determines whether or not each individual is a crossover target from the individuals created by the individual initialization unit 27 (step ST1k). Here, each individual becomes an individual to be crossed (nd) with a probability of CROSSOVER% (step ST2k), and becomes an individual that is not a crossover target with a probability of 100-CROSSOVER%.
Since the crossover probability CROSSOVER is 60% from the preconditions described in FIG. 45, the probability that an individual becomes a crossover target individual is 60%, and the probability that the individual does not become a crossover target individual is 40%. As a result of selecting the crossover target individuals with these probabilities, it is assumed that individuals 0 and 2 are selected as crossover targets. That is, nd = 2.

Next, the two-individual determination unit 41 selects two individuals from the cross-target individual determined by the cross-target individual determination unit 40 (step ST3k). In the above example, since there are only two crossing target individuals, the individual 0 and the individual 2 are selected.
Thereafter, the crossing position determining unit 42 determines the number of crossings between the two individuals selected by the two individual determining unit 41 (step ST4k). Here, since the character string arrangement start candidate point (arrangement start position) to be exchanged between individuals is specified as the number of intersections between two individuals, the number of intersections is referred to as “crossover position” for convenience.
As shown in FIG. 44, the determination method of the crossover position is data in which the individual character string arrangement start candidate points are arranged in order from the origin, so that the individual is registered from the first individual (registered as the crossover target individual). The number of character string arrangement start candidate points is -1) One character string arrangement start candidate point is determined with equal probability among the first character string arrangement start candidate points.
In the above example, the number of character string arrangement start candidate points registered in the individual 0 and the individual 2 is two, so the crossing position is 1.

  Next, the two-individual crossover unit 43 executes a crossover process in which the character string arrangement start candidate points are replaced between the two individuals selected by the two-individual determination unit 41 from the crossover position determined by the crossover position determination unit 42. (Step ST5k). That is, the character string arrangement start candidate points at the cross position from the beginning of the two individuals are interchanged. In the above example, since the crossing position is 1, the first character string arrangement start candidate points of the individual 0 and the individual 2 are switched. By doing so, it becomes as shown in FIG.

  Thereafter, the crossed individuals are excluded from the crossover target individuals (step ST6k), and the processes after step ST3k are repeated until the crossover processing is executed for all crossover target individuals (step ST7k). In the above example, since there are two crossing target individuals, the processing is ended after the above crossing.

(4A) Mutation Processing Here, the mutation processing in step ST10h in FIG. 45 will be described in detail.
In step ST10h, the mutation unit 33 performs a so-called mutation process in which a character string arrangement start candidate point of some individuals is converted into another character string arrangement start candidate point.
In the mutation method, a character string arrangement start candidate point registered in an individual is converted to another character string arrangement start candidate point with a probability of Mutation%. When there are a plurality of other character string arrangement start candidate points, one of them is selected with equal probability.
In the example described above, as shown in FIG. 55, the character string arrangement start candidate point Q1 corresponding to the character string 0 of the individual 2 is converted to another character string arrangement start candidate point by mutation. In the individuals 0 to 2, the character string arrangement start candidate point corresponding to the character string 0 is only Q0 except for Q1, so that Q1 of the individual 1 is changed to Q0.

  After performing the mutation process, the cost function value calculation unit 28 for each individual calculates the cost function of each individual as described above. In the above example, the value of the cost function f of each individual is as shown in FIG. In the cost function value minimum individual storage unit 29, the individual having the lowest cost function value is stored. In the example of FIG. 56, the individuals 0 and 1 are individuals with the lowest cost function values. Thereafter, the convergence determination unit 30 determines whether or not it has converged. In the above example, since two individuals among the three individuals have the minimum value 82 + √244 of the cost function f value, it is determined that there is convergence. Thereby, the process ends.

  When the character string arrangement device 1A according to the second embodiment arranges a character string on the screen, the character string arrangement start candidate point set stored in the output character string arrangement start candidate point storage unit 5a is used. Character strings can be arranged at positions where there is little overlap between columns or between character strings and symbols. In the above example, the set of character string arrangement start candidate points {Q0, Q5} is used for character string arrangement.

As described above, according to the second embodiment, the symbol data acquisition unit 26 that acquires the data of the symbols arranged on the display screen is provided, and the cost function value calculation unit 28 for each individual uses the character string arrangement start candidate. Character strings are arranged so that readings are forward in opposite directions on the display screen starting from the position candidates created by the point creation unit 3, and the positional relationship between the character strings in each arrangement and the symbol data acquisition unit 26. Factors affecting the readability of the character string based on the positional relationship between the symbol and the character string acquired in 26 (duplication of character strings, angle formed by road links, input character string arrangement start point and character string arrangement start A cost function value for evaluating the distance to the candidate point and the overlap between the character string and the symbol is calculated.
By configuring in this way, it is possible to determine the character string arrangement start candidate point having the smallest cost function value including the function indicating the duplication of the character string and the symbol, and not only the duplication of the character strings but also the character string The character layout can be changed to a position that avoids duplication of symbols.
As a result, even when the orientation of the map changes due to a change in the position of the vehicle as in a car navigation system, it is possible to suppress the display of overlapping character strings or character strings and symbols on the screen. it can. Therefore, the character string can be easily recognized.
In addition, by using GA for the process of determining the combination of character string arrangement start candidate points that will be the minimum cost function value, when the number of character strings increases, characters can be calculated with a smaller amount of computation than in the first embodiment. The arrangement can be changed.

  In the said Embodiment 1 and the said Embodiment 2, although the case where a character string was arrange | positioned along a road was demonstrated, this invention is not limited to this. For example, a link (line segment) indicating an image of a linear feature may be a link indicating a river or a railroad track, or a link indicating a boundary dividing a certain region. It is possible.

  In the present invention, within the scope of the invention, any combination of the embodiments, or any modification of any component in each embodiment, or omission of any component in each embodiment is possible. .

  The character string arrangement device according to the present invention can arrange the character string on the display screen without deteriorating the readability without causing an increase in the amount of information to be prepared. Suitable for displaying character strings.

  DESCRIPTION OF SYMBOLS 1,1A Character string arrangement | positioning apparatus, 2 Character string data acquisition part, 3 Character string arrangement | positioning start candidate point preparation part, 4 Minimum cost function value calculation part, 5, 5a Output character string arrangement | positioning start candidate point memory | storage part, 6 Candidate point preparation Part, 7 character string upward placement part, 8 character string downward placement part, 9 candidate point deletion part, 10a, 10b character placement candidate point creation part, 11a, 11b start character placement candidate point determination part, 12a upward character placement order determination Part, 13a upward character temporary placement part, 14a, 14b determination processing part, 15a, 15b character position registration part, 16b downward character placement order determination part, 17b downward character temporary placement part, 18 character and road overlap judgment part 19 character interval determination unit, 20 character overlap determination unit, 21, 21a Character string overlap function calculation unit, 22, 22a Road link angle function calculation unit, 23, 23 Distance function calculation unit, 24, 24a Function value addition unit, 25a Character string upward arrangement unit, 25b Character string downward arrangement unit, 25c Character string overlap calculation unit, 26 Symbol data acquisition unit, 27 Individual initialization unit, 28 for each individual Cost function value calculation unit, 29 cost function value minimum individual storage unit, 30 convergence determination unit, 31 individual re-creation unit, 32 crossover unit, 33 mutation unit, 34 symbol and character string overlap function calculation unit, 35 circumscribed rectangle Creation unit, 36 circumscribed rectangle and symbol overlap determining unit, 37 circumscribed rectangle and symbol overlap function calculating unit, 38 individual ordering unit, 39 recreating unit, 40 crossover target individual determining unit, 412 individual determining unit, 42 crossover Position determining unit, 432 individual crossing unit, 100 display screen, 101 road link, 102, 102a, 102b character string, 103, 103a, 1 3b, 103A string arrangement starting point.

Claims (11)

A character string data acquisition unit for acquiring data of a character string to be arranged on the display screen;
A position candidate creation unit for creating a position candidate on the display screen for starting the arrangement of the character string acquired by the character string data acquisition unit;
Positioning the character strings so that readings are forward in opposite directions on the display screen starting from the position candidates created by the position candidate creation unit, and the positional relationship between the character strings of the respective positions A cost function value calculating unit that calculates a cost function value for evaluating a factor that affects the readability of the character string, based on
A character string arrangement device comprising: a storage unit that stores a position candidate having a minimum cost function value as a position candidate when the arrangement start position of the character string is changed. The position candidate creation unit creates a position candidate for starting the arrangement of the character string along a line segment group indicating a linear image displayed on the display screen,
The cost function value calculation unit starts from the position candidate created by the position candidate creation unit,
The character string is arranged on one side of the line segment group so that the reading is forward in one direction on the display screen, and on the other side of the line segment group on the display screen A function that indicates the overlap between the character strings in each of the arrangements, and a function that indicates the angle formed by the line segments of the line segment group. , And a function value indicating the distance between the arrangement start position set for the character string and the position candidate, respectively, and the sum of these values is used to evaluate a factor that affects the readability of the character string. The character string arrangement device according to claim 1, wherein the character string arrangement device is calculated as a cost function value. The position candidate creation unit temporarily arranges the characters in the character string in order so that reading is forward in one direction on the display screen on one side of the line segment group, starting from the position candidate. Then, from the positional relationship between the characters and the characters temporarily provisionally arranged, the first character string arrangement unit for determining whether or not the character string can be arranged for each character, and the other side with the line segment group as a boundary The characters of the character string are temporarily arranged in order so that the reading is in the forward direction in the direction opposite to the one direction on the display screen. A second character string arrangement unit that determines whether or not the character string can be arranged,
3. The character according to claim 2, wherein the cost function value calculation unit calculates the cost function value starting from a position candidate determined to be laid out by the first and second character string arrangement units. Column placement device.   The first and second character string arrangement units are configured to determine whether the character and the line segment group overlap each other, whether a character interval between the sequentially arranged characters is narrower or wider than a predetermined threshold, and the sequential The character string arrangement device according to claim 3, wherein the character string arrangement device determines whether or not the character string can be arranged with the position candidate as a starting point based on whether or not the temporarily arranged characters overlap. A symbol data acquisition unit for acquiring data of symbols arranged on the display screen;
The cost function value calculating unit arranges the character strings so that readings are forward in opposite directions on the display screen starting from the position candidates created by the position candidate creating unit, Calculating a cost function value that evaluates a factor that affects the readability of the character string based on the positional relationship between the character strings and the positional relationship between the symbol acquired by the symbol data acquisition unit and the character string. The character string arrangement device according to claim 1.   6. The position candidate of the minimum cost function value is determined from among the position candidates for each of the character strings created by the position candidate creation unit using a genetic algorithm. The described character string arrangement device. For the combination of position candidates for each character string created by the position candidate creation unit, ordering is performed according to the size of the cost function value calculated by the cost function value calculation unit, and according to the order of the combination A re-creation unit that creates a new combination of the position candidates for which a cost function value is to be calculated based on the probability distribution created
From the combination created by the re-creation unit, a crossover unit that selects a crossover target combination with a predetermined crossover probability, and performs crossover to replace position candidates between the selected combinations;
The combination created by the re-creating unit and the position candidate of the combination subjected to the crossing by the crossing unit are selected from the position candidates registered in these combinations with a predetermined mutation probability. A mutation part for converting to a position candidate point,
The cost function value calculation unit is a combination created by the re-creation unit until the cost function value of the combination of position candidates for each character string converges to a minimum, a combination in which crossing is performed by the crossing unit, and The character string arrangement device according to claim 6, wherein the calculation of the cost function value of the combination in which the process in the mutation unit is executed is repeated. The position candidate creation unit creates a position candidate for starting the arrangement of the character string along a line segment group indicating a linear image displayed on the display screen,
The cost function value calculation unit starts from the position candidate created by the position candidate creation unit,
The character string is arranged on one side of the line segment group so that the reading is forward in one direction on the display screen, and on the other side of the line segment group on the display screen A function that indicates the overlap between the character strings in each of the arrangements, and a function that indicates the angle formed by the line segments of the line segment group. , A function indicating the distance between the arrangement start position set in the character string and the position candidate, and a value of a function indicating the overlap between the symbol and the character string, respectively, and adding these values together, 6. The character string arrangement device according to claim 5, wherein a cost function value for evaluating a factor affecting the readability of the character string is calculated.   The cost function value calculation unit includes two adjacent characters of the character string arranged in one direction on the display screen on one side with the line segment group as a boundary, and the line segment group. The two adjacent characters in the character string arranged so that the reading is in the forward direction in the direction opposite to the one direction on the display screen on the other side of the boundary through the line segment of the line segment group 8. A character string arranging device according to claim 7, wherein a circumscribed rectangle circumscribing four characters composed of two opposing characters is created, and a function value indicating an overlap between the circumscribed rectangle and the symbol is calculated. In the character string arrangement position candidate determination method of the character string arrangement device including the character string data acquisition unit, the position candidate creation unit, and the cost function value calculation unit,
The character string data obtaining unit obtains character string data to be arranged on the display screen;
The position candidate creation unit creates a position candidate on the display screen for starting the arrangement of the character string acquired in the character string data acquisition step; and
The cost function value calculation unit arranges the character strings so that readings are forward in opposite directions on the display screen starting from the position candidates created in the position candidate creation step, A cost function value calculating step for calculating a cost function value for evaluating a factor that affects the readability of the character string based on the positional relationship between the character strings of the arrangement;
A method for determining a character string arrangement candidate, wherein the position candidate having the smallest cost function value is used as a position candidate for changing the arrangement start position of the character string. A character string data acquisition unit for acquiring data of a character string to be arranged on the display screen;
A position candidate creation unit for creating a position candidate on the display screen for starting arrangement of the character string acquired by the character string data acquisition unit;
Positioning the character strings so that readings are forward in opposite directions on the display screen starting from the position candidates created by the position candidate creation unit, and the positional relationship between the character strings of the respective positions A cost function value calculating unit that calculates a cost function value that evaluates a factor that affects the readability of the character string, based on
A character string arrangement candidate determination program for causing a computer to function as a storage unit that stores a position candidate having the smallest cost function value as a position candidate when changing the arrangement start position of the character string.

Images