JPH04267477A - Range retrieving system by mesh coding of address - Google Patents

Abstract

PURPOSE:To facilitate the retrieve processing of address data in an arbitrary range with the quantity of data and to prevent the retrieving processing and the address data from being affected even when the names of cities, towns and villages are changed, with the geographical two-dimensional data by mesh- coding the address data on a map. CONSTITUTION:All retrieval object areas on the map are divided into equal intervals in an X-Y axis direction and a code is imparted (steps 101 and 102.) The X-axis and Y-axis codes are imparted to each address present in the framework on the map shown by the X-axis and Y--axis codes, the address is mesh- coded and stored(step 103). By the X-axis and Y-axis codes on the map, the retrieval range is designated with an FROM.TO method(step 201). The mesh code in the range shown with the FROM.TO code designated is generated, the mesh code in the range and the mesh code of the stored address are collated with each other and the coincident and corresponding address data are outputted as the retrieved result (steps 202 and 203).

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a range search method for a large amount of address data by mesh-coding addresses existing in a specific range.

0002

[Prior Art] Conventionally, in searching for addresses existing within a specific range, each time a search range is specified, data of all addresses that physically exist within the specified range are collected, and all addresses that are the subject of the search are searched. It was checked against address data.

[0003] Furthermore, whenever a change in the name of a city, town, village, etc. occurs, the search process itself is changed each time.

0004

[Problems to be Solved by the Invention] In the conventional search method described above, it is necessary to have all address data within the search range, and in addition, for one search target data, all address data within the search range must be stored. This had the disadvantage of requiring a large amount of data and complex processing.

[0005] Furthermore, whenever there is a change in the name of a city, town, village, etc., the search process itself must be changed each time, which poses a problem in terms of maintainability.

[0006]

[Means for Solving the Problems] The range search method based on address mesh coding of the present invention includes means for dividing the entire target area into squares at regular intervals in the X-axis and Y-axis directions;
A means for assigning predetermined codes to the divided X-axis and Y-axis for each division unit, and a means for assigning predetermined codes to the divided X-axis and Y-axis for each division unit, and a means for assigning a predetermined code to each of the divided Means for assigning the codes on the X and Y axes as mesh codes, and a means for expressing the range to be searched in the target area in the form of a square or rectangle at the time of search, and assigning the mesh code of the square to which the upper left and lower right vertices belong. means for specifying as a search range; means for generating and outputting the given mesh code included in the search range specified by the two points; and mesh code within the search range that is outputted. and means for comparing the assigned address mesh code to be searched and outputting address data corresponding to an address mesh code that matches any one of all the mesh codes within the search range.

[0007]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be explained with reference to the drawings.

FIG. 1 is a processing flowchart showing one embodiment of the present invention. The present invention can be broadly divided into mesh coding of addresses in search target areas on a map (step 100).
), a search range is specified, and an address is collated and extracted using a mesh code within the specified range (step 200).

Mesh coding of address (step 100)
) divides the entire search target area on the map into equal intervals in the X and Y axis directions (step 101). Next, arbitrary codes are assigned to each of the X-axis and Y-axis directions (step 102
). Next, the X-axis and Y-axis codes are assigned to each address existing within the box on the map indicated by the divided X-axis and Y-axis codes, and the addresses are converted into mesh codes and stored ( Step 103).

[0010] Address matching and extraction (step 200) involves specifying the start and end points of the search range using the X-axis code and Y-axis code on the map to which division codes have been assigned, that is, F
The range is designated by code using the ROM/TO designation method (step 201). Next, a mesh code within the range indicated by the specified FROM/TO code is generated (step 202). At this time, if the specified range is large, a plurality of mesh codes will be generated. Next, match the generated mesh code within the range with the mesh code of the stored address, extract the mesh code of the address that matches the mesh code within the range, and correspond to the mesh code of the extracted address. address data is output as a search result (step 203).

[0011] This will be further explained using FIG. 2 and subsequent figures. As shown in Figure 2, the target area on the map is divided into 5 areas, for example.
Divide into 20 in the X-axis direction and 15 in the Y-axis direction in km units,
Each is given an alphabetic code from A to T and from A to O. Next, as shown in Figure 3, the large divided area is
It is divided into 10 units of 500 m in the axial direction, and each is assigned a numerical code from 0 to 9. Therefore, for example, the mesh code in the shaded area in FIG. 3 is "DH36".

[0012] This mesh code is assigned to each address existing in the box indicated by the mesh code obtained in this way, and the address is converted into a mesh code and stored in, for example, a storage device (not shown). put. The storage structure of the address mesh code is, for example, as shown in Figure 4.
code and 500m code 1 are stored in the X-axis 2 and Y-axis 3 frames, respectively.

[0013] Also, when searching, in order to avoid unnecessary searches due to invalid mesh codes for addresses that cannot actually exist, such as locations on the sea, a check code table (as shown in the figure) is used to check mesh codes that cannot actually exist. ).

Next, search will be explained. Let us take as an example a case where an address existing within a search range indicated by diagonal lines on a divided map as shown in FIG. 5 is to be searched. First, search 5km
A method of generating a unit mesh code (hereinafter referred to as a 5km mesh code) will be explained using FIGS. 5 and 6. In FIG. 5, the mesh code "DA66" at the upper left corner of the diagonally shaded rectangle that is the search range is set as the FROM code, and the mesh code "AE33" at the lower right corner is set as the TO code.

Next, the 5km Y-axis code 1 of the FROM code
Find the letter "D" of 1 in the Y-axis code table 15 of FIG. 6, and from that position until the letter "A" of the 5km Y-axis code 13 of the TO code of FIG. 5 appears on the Y-axis code table 15 of FIG. All character codes are output to the Y-axis code output table 17, and the number of output codes is set to the Y-axis code output counter 21.

Similar processing is also performed for the 5km X-axis code 12 of the FROM code and the 5km X-axis code 14 of the TO code, and from the X-axis code table 16 "A
” to “E” to the X-axis code output table 1
8 and set the output number in the X-axis code output counter 22.

Furthermore, the codes on the generated X-axis code output table 17 and Y-axis code output table 18 are
The codes are combined one by one, outputted to the 5km code output table 20, and the number of outputs is set in the 5km code output counter 23. However, if the combined code at that time matches any of the codes in the check code table 19, it will not be output to the 5km code output table 20 and will not be added to the 5km code output counter 23 as an invalid mesh code.

Next, a method of generating a 500m code will be explained. Since the processing of this 500m code generation method differs depending on the mode of range specification, each method will be explained separately.

FIGS. 7(a), 7(b), 7(c), and 7(d) are diagrams showing examples of range designation, respectively. The difference in aspects is 5k
Y-axis code output counter 2 in Figure 6 when m code is generated
1 and the value of the X-axis code output counter 226.

That is, FIG. 7(a) shows that when the values of the Y-axis code output counter 21 and the X-axis code output counter 22 are both "1", it is the 1 square type 24, and the same , the value of the Y-axis code output counter 21 is "
1'' and the value of the X-axis code output counter 22 is ``2 or more'', it indicates the horizontal type 25, and (c) indicates that the value of the Y-axis code output counter 21 is ``2 or more'' and the When the value of the axis code output counter 22 is "1", it indicates that the vertical type 26 is used. ” indicates that it is other type 27.

Next, each process will be explained in detail. The 1-cell type 24 simply combines the 500m codes of the FROM and TO codes and outputs them to the 500m code 35 of the search table shown in FIG. . However, all other types of processing ultimately generate a mesh code for each square, and this single square type processing is commonly used.

FIG. 9 is a flowchart showing the process for landscape type 25. The processing counter has an initial value of 0.
” is given (step 35), and then the processing counter is set to 1.
5 of the search table shown in FIG.
5km obtained from the 5km code output table 20 in FIG. 6 by adding the value of the processing counter to the km code 28 as the subscript value.
Transport the code (step 37). Furthermore, the 500m code of the FROM code specifying the search range in FIG. 5 is transferred to the 500m FROM code 29 of the mesh code in FIG. 8 (step 38). Then, the value of the processing counter is determined (step 39), and if the value is "1", processing 1 is performed (step 40), and the process returns to step 36. The value is "1"
If not, the value of the processing counter is judged again (step 41), and if it is smaller than the value of the X-axis code output counter 22 shown in FIG. 6, processing 2 is performed (step 42), and the process returns to step 36. If the value is not smaller than the X-axis code output counter 22, process 3 is performed (step 43), and the process ends.

Process 1, which will be discussed here, transfers "9" to the TO code X axis 34 of the search table shown in FIG.
moves "0" to the FROM code X-axis 32 of the search table and "9" to the TO code X-axis 34, and process 3 moves "0" to the FROM code X-axis 34 of the search table. That is, as shown in FIG. 10, a search mesh code is generated for each cell.

FIG. 11 is a flowchart showing the process for portrait type 26. However, the overall process flow is the same as the horizontal type process, so the process 1, which has different processes,
Only Process 2 and Process 3 will be explained.

In processing 1 here, "9" is set on the TO code Y-axis 33 of the search table shown in FIG. "9" is the FR of the search table in process 3.
“0” is transferred to the OM code Y-axis 31, respectively. That is, as shown in FIG. 12, a search mesh code is generated for each cell.

FIG. 13 is a flowchart showing processing of other type 27. This type of processing uses a work area to change the contents of the FROM code that specifies the search range, and then changes the work area as if it were FROM/T.
It treats it like an O code and performs horizontal type processing, and finally generates a 500m mesh code.
) (b) shows and explains the data structure of each work area.

First, the processing counter has an initial value of "1".
” is given (step 53), and F shown in FIG. 14(a)
The FROM code with the search range specified in FIG. 5 is transferred to the ROM code work area 70 (step 54).
, TO code work area 7 also shown in Figure 14(b)
5, transfer the TO code in Figure 5 as it is (Step 5
5). Next, determine the value of the processing counter (step 56).
, if the value is larger than the value of the Y-axis code output counter 21 shown in FIG.
In the OM code work area 5km code 66, set the processing counter 1 from the 5km code output table 20 shown in FIG.
A 5 km code obtained by adding 1 to the value obtained by multiplying the subtracted value by the value of the X-axis code output counter 22 shown in FIG. 6 is transferred (step 57).

Furthermore, in the TO code work area 5km code 71 shown in FIG. 14(b), the value of the processing counter and the X shown in FIG.
The 5km code obtained by multiplying the value of the axis code output counter 22 by using the subscript value is transferred (step 58).

Then, the value of the processing counter is determined (step 59), and if the value is "1", processing 1 is performed (step 60), and the process proceeds to step 64. If the value is not "1", the value of the processing counter is judged again (step 61), and if it is the same as the value of the Y-axis code output counter 21 shown in FIG. 6, processing 2 is performed (step 62), and the process proceeds to step 64. ,
If they are not the same, process 3 is performed (step 63). Here, the FROM code work area 70 and TO code work area 75 in FIG. 14 processed by processes 1 to 3
FROM code and TO as shown in Figure 5, respectively.
As a code, perform the horizontal type 500m code generation process shown in FIG. 9 to generate a 500m code (step 64), and add 1 to the processing counter value (step
5), return to step 56.

Processing 1 described here transfers "9" to the TO code work area 500m code Y-axis 73 in FIG. 14, and processing 2 transfers "0" to the FROM code work area 500m code Y-axis 68. 3 transfers "0" to the FROM code work area 500m code Y-axis 68 and "9" to the TO code work area 500m code Y-axis 73.

That is, as shown in FIG. 15, the entire search range is divided into horizontal types, and a mesh code for searching is generated for each square by performing horizontal type processing as shown in FIG. It turns out.

Through the above processing, all mesh codes within the search range are generated in the format shown in FIG. 8 in the above embodiment, and are compared with the address mesh code to be searched to find address data corresponding to the same mesh code. Output.

That is, for one search target address mesh code, it is determined whether the 5km code 1 shown in FIG. 4 is the same as the 5km code 28 of the search mesh code shown in FIG. The 500m code Y-axis 2 of 4 is between the FROM code Y-axis 31 and the TO code Y-axis 33 of FIG. 8, and the 500m code X-axis 3 of FIG. It is determined whether the target address data is between the axes 34, and if it is between them, it means that the target address data exists within the search range, and if it is not between them, the data is outside the search range. In addition, if the 5km codes are not the same, the above processing is performed with the next search mesh code to determine whether the data is within the search range, data outside the search range, or the search mesh code is changed. Repeat until you have referenced everything.

[0034] By performing this for all address data items to be searched, the search processing is completed.

[0035] Furthermore, in contrast to the embodiments described so far,
Depending on the size of the entire area to be searched and the contents of the search target, the mesh code assignment method, especially the division unit X
The way the axis and Y-axis codes are given differs. If the number of divisions is small, a two-stage mesh code such as the generation of the 5km code and 500m code as described above would be unnecessary, and a single-stage mesh code would be sufficient. If the data is converted into numerical data, the process of generating the 5km code as described above is not necessary.

[0036]

[Effects of the Invention] As explained above, the present invention has the advantage that by managing address data as two-dimensional data called mesh codes, it is possible to easily search for address data in any range using a small amount of data. be.

Furthermore, since addresses are treated as two-dimensional geographical data, search processing and address data are not affected in any way even if the name of a city, town, village, etc. is changed, resulting in high maintainability.

[Brief explanation of the drawing]

FIG. 1 is a processing flowchart showing one embodiment of the present invention.

FIG. 2 is a diagram showing an example of dividing a search target area into 5 km units.

FIG. 3 is a diagram showing an example of dividing a search target area into units of 500 m.

FIG. 4 is a diagram showing an example of an address mesh code.

[Figure 5] (a) to (c) are search range FROM/TO
FIG. 3 is a diagram showing an example of code specification.

FIG. 6 is a diagram showing an example of generating a 5km code, which is a large division unit.

FIGS. 7(a) to 7(d) are diagrams of how the search range is specified.

FIG. 8 is a diagram showing a search table that stores search mesh codes.

FIG. 9 is a flowchart illustrating 500m code generation processing for horizontally long type search range specification.

FIG. 10 is a flowchart illustrating a mesh code generation process for each cell in landscape type processing.

FIG. 11 is a flowchart illustrating search range specified vertical type 500m code generation processing.

FIG. 12 is a flowchart illustrating a mesh code generation process for one square unit in portrait type processing.

FIG. 13 is a flowchart showing search range specification and other type 500m code generation processing.

FIGS. 14(a) and 14(b) are diagrams showing an example of the structure of a FROM code work area and a TO code work area.

[Figure 15] Other types of processing, landscape type processing, and 1
It is a figure explaining the relationship with mesh code generation processing per cell.

[Explanation of symbols]

100 Mesh encoding of addresses in the search target area on the map Step 101 Division processing step 10 for all search target areas
2. Coding process step 103 in division units
Address mesh encoding processing step 200
Matching and outputting address using mesh code step 201 Search range specification processing step 202
Specified range mesh code generation processing step 203 Collation output step of specified range mesh code and search target data

Claims (2)

[Claims] Claim 1: In a range search method using address mesh coding to search for address data existing within a specific range within a target area, the entire target area is divided into squares at regular intervals in the X-axis and Y-axis directions. means for dividing the divided X-axis and Y-axis into each division unit, and means for assigning a predetermined code to the divided X-axis and Y-axis for each division unit; Means for assigning the code on the X-axis and Y-axis of the square where the square actually exists as a mesh code, and when searching, the range to be searched in the target area is expressed in the form of a square or rectangle, and the upper left and lower right vertices of the range are expressed as a mesh code. means for specifying the mesh code of the square to which the cell belongs as a search range; means for generating a mesh code within the search range that generates and outputs the given mesh code included in the search range specified by the two points; means for comparing the given mesh code within the search range with the assigned address mesh code to be searched and outputting address data corresponding to an address mesh code that matches any of all the mesh codes within the search range; A range search method using mesh coding of addresses, characterized by the following. 2. The within-search-range mesh code generation means uses a check code table stored in advance as a check code for mesh codes whose addresses cannot exist among the mesh codes, and generates mesh codes within the search range. 2. A range search method based on mesh coding of an address according to claim 1, further comprising means for excluding and outputting a mesh code corresponding to said check code.