JP4121966B2 - Graphic data management method - Google Patents

Description

  The present invention relates to a graphic data management method in a graphic processing system such as GIS (Giographic Information System) or Computer Aided Design (CAD), and changes in a town due to new construction or removal of a house. It is related with the efficient management method of the data which changes with time like.

In business using geographic information in a local government or the like, it is important to store past history information. In facility management, it is necessary to grasp the maintenance information and usage status so far, and in city planning, the transition of the city so far is useful for analysis of the current situation and prediction of the future.
Until now, local governments have managed this geographic information on a paper map, (1) changes and changes were made directly to the map, (2) the entire map was rewritten once every few years, and (3) past paper maps. Used to be stored in the warehouse. However, in this method, searching past maps is a difficult task, and storing a large amount of past paper maps has become a burden.
In recent years, GIS (Geographic Information System) based on digitization of paper maps has been developed and used. Since GIS handles digitized maps, it is relatively easy to update maps and store and retrieve past maps compared to paper maps.
In order to store the past geographic information in the conventional GIS, the entire content of the geographic information DB (Data Base) is regularly backed up to an external storage medium such as DAT (Digital Audio Tape) once a month. I was taking.
Further, as a map use system, there are a map database management method when map data is deleted and added, and map data is collectively changed. (For example, see Patent Document 1)
However, in these methods, (1) when searching in the time axis direction, for example, a search “changes in this region from the present to 10 years ago”, it is necessary to refer to all of the past backups. (2) Since all geographic data is subject to backup, backup time and storage space for historical data are always a heavy burden.
With the spread of the Internet and intranet, GIS network support has been performed. Geographic information is stored in the geographic information DB server, and each user can access the server from his / her terminal and acquire a map and other geographic information via the network. However, since geographic information generally has a large amount of data, (3) there is a problem of increased communication time and increased load on the network.

JP-A-8-44846

An object of the present invention is to solve the following three problems.
(1) History information management that makes searching in the time axis direction fast and easy.
(2) History information management that minimizes maintenance (backup) time and history data storage space.
(3) A communication method that minimizes the communication time and the load on the network when acquiring maps and other geographical information via the network.

In order to achieve the above object, the present invention provides:
It is a graphic data management method in which each of a plurality of features such as houses and roads is used as graphic data, and these graphic data are registered and managed in a database.
Provide the feature generation time and disappearance time data as the feature graphic data,
If a feature changes over time, for a newly generated feature, the generation time is stored in the graphic data of the feature and registered in the database. Store the extinction time in the graphic data of the feature registered in the database and register it in the database,
When the graphic data of a feature registered in the database is searched by specifying a time, the graphic data of the feature whose specified time falls between the generation time and the disappearance time can be acquired. .

In addition, each of a plurality of features such as houses and roads is used as graphic data, and the graphic data is registered and managed in a database.
As the graphic data of the feature, a set of data including a change time of the feature, a change attribute, and a change difference of the attribute is provided.
When a change occurs in the feature with the passage of time, the change time changes in the set of the graphic data of the feature registered in the database in accordance with the change every time the change occurs. The time when the change occurred, the change attribute of the attribute of the graphic attribute is changed, the change amount of the attribute is stored in the attribute change difference, and the graphic data of the feature is registered in the database,
When the feature graphic data registered in the database is searched by specifying the time, the feature graphic data including the change up to the specified time can be acquired.

In addition, a unique ID number is given to the graphic data of each feature,
ID number data of past graphic data is provided as additional information of the generation time of the graphic data, ID number data of future graphic data is provided as additional information of the disappearance time,
When a change occurs in the feature over time, the figure data of the changed feature is created, and the ID number data of the feature data before the change is used as the ID number data of the previous figure data of the figure data. Storing the ID number of the graphic data of the feature after the change as the ID number data of the graphic data of the future of the graphic data of the feature before the change, and registering each in the database,
It makes it possible to manage changes in features over time.

In addition, each of a plurality of features such as houses and roads is made as graphic data, a set of figures is set as an area, and the graphic data is registered and managed in the database as graphic data belonging to the area.
A huge figure extending between the areas is represented by a broken line, and the figure ID and area ID of the connection figure and the figure ID and area ID of the end point are stored in the figure data belonging to each area of the giant figure. The graphic data is registered in a database so that huge features exceeding the area can be managed.

  When transferring graphic data to a client system connected to the database via a network, if the client system has already acquired old graphic data in the database, the old graphic data in the database The graphic data change difference at the designated time is transferred to the client system.

  If the database does not require a time search and a high-speed spatial search is required, only the graphic data existing at the specified time is extracted from the database, and the extracted graphic data The time data is removed from the database, the database from which the time data is temporarily removed is reconstructed, and high-speed spatial search is performed on the reconstructed database.

Further, by setting the rotation angle, the parallel movement amount, and the enlargement ratio of the figure as the change attribute, it is possible to express the change of the feature with a small number of parameters.
When transferring graphic data acquired in the past by the client system to a client system connected to the database via a network, the graphic data acquired in the past and the graphic at a specified time The change difference from the data is transferred to the client system.

This method has two effects.
First, this method reduces the amount of geographic information data.
If the size of the geographic information that does not include time information is C, the geographic information has been updated n times so far, and the percentage of the total geographic information changed per update is p, the current size of the geographic information Is (1 + n) C in the conventional method for backing up the entire geographic information every time it is updated, and (1 + pn) C in the present method.
Assuming that the number of updates n = 100 and the ratio p = 1%, this method has a data size of about 1/50 that of the conventional method.
Secondly, this method can extract geographic information at any time.
In the conventional method, the update interval is a long interval such as once a year because of the problem of the data amount described above, and the situation within the interval cannot be known.
In this method, the extraction of geographic information at an arbitrary time is realized by adding time information indicating when each update target has changed at the time of update.

Embodiments of the present invention will be described below with reference to the drawings.
FIG. 2 is a configuration diagram for realizing a geographic information system, which is an embodiment using the time history management method of the present invention.
The display 201 is used for displaying graphics. The keyboard 202 and the mouse 203 are interface devices used for designating graphics when searching / editing graphics.
The printer 204 is used for graphic output. The scanner 205 is used to read and digitize a figure such as a paper map.
Digitized figures are stored in a geographic information DB (Data Base) 209 in a storage device such as a memory / hard disk.
The geographic information system is executed on a computer 208 such as a personal computer or a workstation.

Further, the geographic information system can be distributed on a network 210 such as a LAN (Local Area Network).
The user client 208 for retrieving geographic information and the server 212 having the geographic information DB are mounted on different computers, and the geographic information in the server can be retrieved from a remote location.
Furthermore, the geographic information DB 211 can be managed by distributing to a plurality of servers by region and field, and one client can access a plurality of servers.
The geographic information system has access to a local geographic information DB 209 in its own system and another geographic information DB 211 on the network, and searches / updates the geographic information, and a plurality of application units 208 operating thereon. Composed.
The application section includes a geographic information reference / update tool, a tool for creating geographic information from a paper map, aerial photographs, and the like, and a tool for performing analysis / simulation using geographic information.

FIG. 3 and FIG. 4 show the data structure of geographic information that is the basis of the expansion of the present invention.
Geographic information is composed of a figure 302, a component point 303, a region 301, and attribute information 304.
A figure 302 represents the shape of a feature such as a house or a road.
A figure is basically represented by a polygonal line. For example, roads are represented by open lines, block boundaries and contour lines are closed lines, houses are polygonal columns obtained by adding height to closed lines, and underground pipes are represented by three-dimensional expanded lines.
401 shows the data structure of the figure.
The figure is composed of its ID, attributes such as line color / number of constituent points, and one or more constituent points 303. One point (305) can also be seen as a figure with its constituent points.
As the graphic ID, a serial number unique to the entire system is used. In order to maintain uniqueness between different systems, the coordinates of the figure (such as the center of gravity) can be substituted.

The configuration point 303 is a point that forms a polygonal line of the figure 302. 402 shows the data structure of the component points.
A constituent point is composed of attributes such as the color of the point, and coordinates x and y (including z in the case of a three-dimensional figure) where the point is located.

An area 301 indicates an area including a figure. This corresponds to the leaf on the paper map.
The shape of the area is arbitrary. For example, a 500 m × 700 m rectangular mesh or an irregularly shaped block can be used as the area.
Reference numeral 403 shows the data structure of the area.
The area is composed of its ID, what kind of geographic information it is (road map, city planning map, etc.), attributes such as when it was created, and zero or more figures 302 included in the area.

The attribute information 304 is information other than graphics linked to geographic information.
For example, when a house figure is added with the site area or the name of a householder, and a road width, road type, traffic regulation information, etc. are added to a road figure, they are managed as attribute information. Arbitrary information such as documents, images, and other figures can be added to the attribute information.
The attribute information stores the ID of the attribute information in the attribute area (shown as an attribute in the figure) of the figure 401, the component point 402, and the area 403 constituting the geographic information, and refers to the attribute information 404 in another table. Can be managed.
404 shows the data structure of attribute information.
The attribute information includes the ID and the content of the attribute.

In the following, time expansion performed on these pieces of geographic information will be described.
In the present invention, the time extension is classified into three extensions: (1) “generation / annihilation” extension, (2) “change” extension, and (3) “replacement” extension.
In the following, these time extensions will be described with reference to graphics constituting geographic information. However, similar time extensions are possible for the constituent points, areas, and attribute information constituting the geographic information.

“Generation / Disappearance” extension refers to the generation / disappearance of geographic information, for example, when a house is newly built and demolished.
FIG. 5 is an explanatory diagram of the “generation / annihilation” extension.
Consider a house that was newly built with 501, 502, and 503 over time lapses of time T 1, T 2, and T 3 and was demolished.
It is assumed that only the building C exists at 501, the building A is newly built at 502, the building B is newly built at 503, and the building C is demolished and removed.

Conventional history management (FIG. 18) has been realized by storing all geographic information (graphics) at a certain point in time.
Building C (1804) was stored as geographic information (1801) at time T1, houses A and C (1805) were stored at time T2 (1802), and houses A and B (1806) were stored at time T3 (1803).

In the present invention, the generation time Ts (508) and the disappearance time Te (509) are added to the graphic data structure 507.
When the generation / disappearance of the graphics A, B, and C is expressed on the time axis, 504 for A, 505 for B, and 506 for C.
When a time search such as “Show the year, month and day of the geographic information” is performed on this geographic information, the target time is compared with the generation / extinction time of the figure, and the target time is generated / It can be determined that the geographical information exists if it is during the extinction time and does not exist otherwise.
When the figure does not disappear and still exists, the identifier (∞) of “annihilation time undecided” is stored in the disappearance time.

The “change” extension indicates a change in geographical information, for example, the height of a house is reconstructed from a one-story building to a two-story building.
FIG. 6 is an explanatory diagram of the “change” extension.
It is assumed that the building C is generated at the time Ts, the attribute changes at the time Tc (for example, the first floor changes to the second floor), and disappears at the time Te.
In order to express the “change” extension by expanding the data structure (507 in FIG. 5) expanded to “generation / annihilation”, the generation time Ts (608), the extinction time Te ( 609), three pieces of change information of a change time 610, a change attribute 611, and an attribute change difference 612 are inserted.
The change time 610 describes the time Tc when the change occurred. A change attribute 611 and an attribute change difference 612 form a set, which indicates what kind of attribute of the graphic has changed and what value has changed.
The three pieces of change information are inserted into the data structure each time a change occurs.
The number of times of change is described in the header information of the graphic.
When a plurality of attributes change at the same time Tc, a set of a plurality of change attributes 611 and attribute change differences 612 per change time 610 can be stored as an abbreviated expression.
The attribute of the figure at a certain search time is obtained by accumulating the change difference up to this search time in the initial attribute at the time of figure generation.
The “change” extension changes the attributes of the graphic, so that the number of types can be defined as many as the number of attributes.

FIG. 7 shows typical attribute changes.
Examples include layer / color change 701, height change 702, and the like. Here, the layer indicates a classification of city boundary lines, house frames, road lines, house floors, and the like.
“Change” expansion is performed by changing the height of the house, changing the material of the house, changing the residents' information, etc.
Also, translation 703, rotation 704, and enlargement / reduction 705 can be defined as changes in the entire figure.
When there is a shift in the position of a figure due to a shift in the position of the house due to an earthquake, relocation, or a change in the city plan, the figure is translated, rotated or scaled.
In these cases, it is necessary to recalculate the constituent points at the time of graphic display / retrieval. However, since the change of the graphic can be expressed with a small number of parameters, the capacity of the entire geographic information can be reduced.
The attribute change difference 612 stores a figure movement amount (Δx, Δy) for parallel movement, a rotation amount θ for rotation, and an enlargement ratio r for enlargement / reduction.

Similar to the “change” extension, the “replacement” extension also indicates a change in geographic information.
The “change” extension has change information in the figure, but the “replacement” extension uses two or more figures and expresses the change by replacing the figure with another figure.
The “replacement” extension is used to describe a change that is too large to be represented by the “change” extension and the data size becomes large.
Also, if the same figure repeats “change” expansion, the figure itself will increase in size and take a long time to search. However, this can be divided into multiple figures by “replacement” expansion to improve the search speed. it can.

FIG. 8 is an explanatory diagram of the “replacement” extension.
Assume that the building C is rebuilt to the building B at the time Ts and is rebuilt to the building A at the time Te.
The figures of building A, building B, and building C are stored in the area as separate figures.
In order to express the “replacement” extension by extending the data structure (507 in FIG. 5) expanded to “generation / annihilation”, the generation time Ts (804), the extinction time Te ( 806) As each incidental information, ID 805 indicating the past graphic 810 and ID 807 indicating the future graphic 808 are stored.
If the ID field of the past graphic / future graphic is empty, it is defined that the graphic is generated / disappeared. If the ID field is not empty, it is defined that the graphic is replaced with the graphic indicated by the ID.
If the figure does not disappear and still exists, the identifier (∞) of “determination time undecided” is stored in the disappearance time.

FIG. 1 shows a general-purpose table to which all of the three time extensions of graphics described above, (1) “generation / extinction” extension, (2) “change” extension, and (3) “replacement” extension are added.
Here, the pre-generation graphic ID is the same as the past graphic ID, and the post-annihilation graphic ID is the same as the future graphic ID. In addition, there are as many sets of change time, change attribute, and attribute change difference as there are changes, and there are as many component point coordinates as there are component points.
This is a time extension taking a figure as an example, but the same table extension is also applied to other geographic information and constituent point / region / attribute information.
The upper part of the figure schematically shows that (1) search in the time direction can be performed, (2) backup of only the updated data can be performed, and (3) only the latest data can be transmitted.
(1) Search in the time direction is to add time information to a figure, and to search a map or information in a spatial region at any time in the past and any time in the future at high speed. ) Backup of only updated data is that, for example, only information older than 10 years can be selected and backed up to the external storage device. (3) Only the latest data can be transmitted, for example, by the client up to 1 year ago. If so, it is only necessary to transmit the latest data thereafter.

FIG. 9 shows an example of time extension applied to the component points in the same way as a figure.
In this time extension, a “generation / annihilation” extension over time of a component point is expressed.
This is realized by adding the generation time Tstart (905) and the extinction time Tend (906) to each component point information 904 in the figure data structure 901, as in the data structure 902, to form a set of information. To do.
The generation / disappearance of the component points can be used when, for example, the sewage pipe is represented as one figure and the position and direction of the pipe is changed by the expansion work.
In addition, it can also be used when the shape (contour lines) of a cliff changes due to landslides, or when the shape of a cliff changes due to the extension of a house.
However, the data structure 902 is a redundant structure having time information at all the configuration points.

Data structure 903 is an extension of data structure 902 and implements data compression.
In the data structure 903, a 1-bit time extension flag is provided at the head of each component point information.
A configuration point whose time extension flag is ON has a generation time and an extinction time, and a configuration point whose time extension flag is OFF has no time information.
The generation time / disappearance time of the component point having no time information follows the generation / disappearance time of the entire graphic stored in the graphic header.
It is assumed that the house 907 (908) represented by the data structure 901 is expanded to a house 909 at time Tstart.
The construction points E, F, and G are newly added to the quadrilateral ABCD to form a polygon ABEFFGCD.
In this case, the configuration points E, F, and G are inserted between the points B and C with the start time and the end time.
The data structure of the table is 903.

FIG. 10 shows another example of time extension applied to the configuration points.
Consider a situation in which a configuration point C1 is added to the figure ABD (1001) at time Tstart (1002), the configuration point C1 changes to C2 at time Tmove (1003), and C2 is deleted at time Tend.
In this time extension, “generation / annihilation” extension and “change” extension are added to the component point information.
The data structure 1005 shows a data structure obtained by extending the data structure 903 and adding “change” extension to the constituent point information.
In addition to the generation time Tstart and the annihilation time Tend, a new set of coordinate values is inserted as zero or more pieces of change information, that is, the change time Tmove and the attribute change difference, with respect to the configuration point C subjected to time extension. Note that “1” and “0” are flags indicating the continuation of the description, “1” indicates that the description continues until the extinction time Tend, and “0” indicates that the description does not continue.

FIG. 11 shows an example of time extension for the area table, similar to the figure.
By providing the area with a time range and dividing the area at a fixed time (for example, in units of one year), it is possible to speed up the search for geographic information.
When searching for geographic information located at a certain point (x, y, z, t) or geographic information that exists within a certain time range from all geographic information, all figures in all areas in the system are squeezed. Instead of searching, the area including the coordinates to be obtained from all the areas is first narrowed in advance in terms of space and time, and the figure is searched for the narrowed area, thereby speeding up the search.

FIG. 11 represents a “replacement” extension of the region.
The generation time 1102 and the annihilation time 1104 are added to the area information 1101, and the ID 1103 of the area before generation is added as incidental information of the generation time 1102, and the ID 1105 of the area after annihilation is added as incidental information of the erasure time 1104.
When this extension is performed, the generation / annihilation time is not the generation / extinction of the area, but the switching time to the next linked area.
As another example of “replacement” expansion of an area, when the total capacity of the geographic information DB increases and does not fit in the system, the area is divided and old geographic information is replaced with another geographic information DB or DAT. It is possible to use such that the external storage device is saved.
Generation or disappearance of an area can be expressed by emptying the link destination ID before or after the creation.
As an example of the generation / disappearance of the area, there may be a case where a block with a certain name A disappears and a block with a new name B is newly created for the purpose of partitioning.
In this case, for the block of name A, the link destination ID after erasure is emptied, and for the block of name B, the link destination ID before generation is emptied, and the link destination ID after annihilation is “destruction time undecided”. Stores the identifier (∞).

FIG. 12 shows an example of time extension for attribute information.
The attribute information is a document / image linked to a graphic (1201) and has a data structure of 1202. Usually, the attribute is managed as an attribute (1203) in a separate table, and the ID of the attribute is stored in the graphic.
Here, for example, when managing attributes (1203) that change with time, such as the number of inhabitants that change every year, time extension for attribute information is used.
The “change” extension shown in FIG. 6 is extended to the graphic information 1202 (1205), the generation time 1208 and the disappearance time 1209 are added, and a set of zero or more change time 1210, change attribute 1211, and attribute ID 1212 is inserted between them. The attribute of the attribute ID 1212 is managed in the table shown as the attribute 1206 in the figure.

FIGS. 13 and 14 show an example in which “replacement” expansion, which is time expansion of geographic information according to the present invention, is applied in the spatial direction.
FIG. 13 shows four areas divided into meshes.
Among the figures stored in the area, there are also huge figures that exist across multiple drawings and areas, such as a national highway.
These figures are divided into two at the boundary line of the region, and the two divided figures are not related as they are.
For example, when performing a shortest distance search using a road network and when a figure must be handled as one, it is necessary to provide some relevance.

As a way to solve this, a “replacement” extension is applied between the spatial domains.
In general, a figure is represented by a polygonal line and has a start point and an end point. Therefore, the graphic table 1301 stores the ID of the graphic connected to the start point and the ID of the graphic connected to the end point. When the figure ends at the start point / end point, the end ID (0) is stored.
A pair of connection area ID and connection figure ID is stored to indicate a figure that crosses the area. The area ID may be automatically calculated as an adjacent area.
In this way, by storing the ID of the graphic following the adjacent area as the connected graphic ID, it is possible to represent a huge graphic across the area.
FIG. 14 shows a management chart of a building having a plurality of floors.
A building can be represented by a combination of floor plans separated by floors.
However, there are huge figures across the floors, such as electrical wiring.
Each floor is divided into each area, the start and end points of the figure are set at the vertical boundary of each floor, and the ID of the figure following the adjacent area is stored as a pair of connection area ID and connection figure ID. It can express a huge figure.

Next, as a maintenance method of the geographic information DB of the present invention for managing the time history, an area time division method will be described.
In the time history management method of the present invention, both past geographic information and the latest geographic information are handled in the same row. Therefore, if past information increases too much, the DB storage space increases, the search time increases, and the performance of the system is reduced. To reduce.
Therefore, when the area is divided into the past “old area” and the “new area” that is close to the present, and when searching for recent changes in the figure, the figure data is searched only from the new area, not from all areas. Efficiency is required.
Similarly, when searching for past figures, if you divide the area in units of one year, etc., search for the area that contains the time range you want to search first, and search for graphic data from that area. Efficient.

In order to divide the area, it is necessary to divide the area information itself and to select and divide the graphic information included in the area information.
Among the figures, there are a figure that is completely included in the time range of the area and a figure that spans the time of area division.
The figure that is completely included in the time range of the area can be selected by selecting which area, but the figure that spans the area division time can be managed by overlapping two areas. It is necessary to manage and divide the figure itself.

The method of selecting and dividing the figures in accordance with the area division is as follows: (1) When dividing an area to improve search speed, (2) When saving or storing past history in backup media, (3) The latest geography on the server It is divided into three cases when requesting information.
FIG. 15 shows a case where the area is divided to improve the search speed.

As described above, it is efficient to divide an area into units of one year or the like, and first obtain an area including a time range to be searched at the time of search, and search for graphic data from the area.
When the region 1501 is divided into the region 1502 and the region 1503 in time Tc, the divided regions 1502 and 1503 must include all the figures included in the respective time ranges. Therefore, a figure that straddles time Tc must be managed in both areas.
The figure A, B, C, D of “Generation / Extinction” extension, the figure E of “Change” extension, and the figure F of “Replacement” extension are divided by the following method.
Of the figures subjected to the “generation / extinction” extension, the figure A that spans between the two areas is redundantly managed in both areas.
Figures B and C existing only in one area are managed only in each area.
In addition, the graphic D that still exists and whose annihilation time is undetermined is redundantly managed in both areas.
The figure E subjected to the “change” extension is divided into two pieces of figure information at the time closest to the division time Tc among the change times.
Since the time search is performed in the vicinity of the division time, the part that crosses the division time is managed by overlapping both figures. The two pieces of graphic information are linked to each other as a graphic subjected to “replacement” expansion.
Since the figure for which the “replacement” extension is performed is a case where a time search is performed in the vicinity of the division time, the figure F that crosses the division time is managed by overlapping both figures. The link destination ID holds the value as it is.

The above management method is redundant because a graphic that crosses the division time is redundantly managed.
In particular, when the time history of geographic information of cities is managed in the order of several years, the figure D, which is the “figure that has not changed since generation” of the figure information of FIG. A lot of storage space is required.
FIG. 16 shows a case where the past history is saved and stored in the backup medium.
When the area 1601 is divided into the area 1602 and the area 1603 at time Tc, the area 1602 closest to the present among the divided areas includes all the diagrams included in the time range.
The area 1603 past the time Tc for saving to the backup medium does not include any figure that crosses the time Tc. As a result, there is no overlapping of figures, and the capacity can be kept to a minimum.

FIG. 16 shows the region division processing.
The figure A, B, C, D of “Generation / Extinction” extension, the figure E of “Change” extension, and the figure F of “Replacement” extension are divided by the following method.
As for the figure that has been “generated / erased” extended, the figure B that exists only in the old area 1603 is managed in the old area, and all the others (A, C, D) are managed in the new area 1602.
The figure E subjected to the “change” extension is divided so that only the part completely existing in the old area 1603 is managed in the old area 1603 in the time range divided by the change time.
Duplicate management is not performed for parts that cross the division time. The two pieces of graphic information are linked to each other as a graphic subjected to “replacement” expansion.
The figure F subjected to the “replacement” extension is selected so that only the part completely existing in the old area 1603 is managed in the old area 1603.
Duplicate management is not performed for parts that cross the division time. The link destination ID holds the value as it is.
When searching for a time past the time Tc, both the area 1602 and the area 1603 are searched.
When searching for a time closer to the present time than the time Tc, only the area 1602 may be searched.

FIG. 17 shows a case where the client requests the latest geographic information from the server.
The present invention is applied to communication between a plurality of geographic information DB servers and clients distributed on a network such as a LAN, thereby realizing reduction in network load and high-speed response.
A function is necessary in which geographic information is stored in a DB server on the network, and a plurality of users refer to and update the geographic information from each client.
For example, when a geographic information system is used in a local government or the like, there are a plurality of departments that maintain and use geographic information, and they are intertwined with each other.
Each section in charge, such as the road section for roads and the water supply section for water and sewage systems, performs maintenance on the features it manages.
In addition, since there is an enormous amount of geographic information, a plurality of persons in charge share the area and perform maintenance.
On the other hand, the capacity of geographic information is generally large. For example, the average three-dimensional geographic information of one area (500 m × 500 m) is about 1 MB. When it becomes four-dimensional geographic information, the capacity further increases.
For this reason, it is a problem to reduce the load on the network and the communication time when downloading and uploading geographic information. This is particularly noticeable when a telephone line with a low transfer rate is used.
In the following, a difference information communication method of geographic information for solving this problem is shown.
Large-capacity four-dimensional geographic information is distributed to clients in the form of a large-capacity storage medium such as CD-ROM or DVD-RAM in advance.
When the user requests the latest geographic information, the latest geographic information is distributed via the network in the form of “difference four-dimensional geographic information”.
In FIG. 17, it is assumed that the client 1702 owns the latest data 1705 at time Tc, and wants to download geographic information updated from time Tc to Ts at the current time Ts.
Also, assume that the server stores graphics A, B, C, and D for “generation / extinction” expansion, graphic E for “change” expansion, and graphic F for “replacement” expansion.
When a geographic information history between Tc and Ts is requested from the client, the server returns update difference information 1704. The update difference information 1704 is only a graphic that has changed between times Tc and Ts.
Of the figures that have undergone the “generation / disappearance” extension, A disappeared between times Tc and Ts and C generated / disappeared become update difference information.
The figure E subjected to the “change” extension is divided by the shortest length including the division time Tc among the change times.
As the figure F subjected to the “replacement” extension, only the figure that has changed between the times Tc to Ts is stored. The link destination ID holds the value as it is.

The client that has received the update difference information 1704 can acquire the latest four-dimensional geographic information 1706 at the current time Ts by adding the update difference information 1704 as it is to the geographic information 1705 that the client has.
In this method, since only the updated geographic information 1704 passes through the network, communication time and network load can be suppressed.
If the client does not need history information and only wants the currently valid geographic information, the client needs to send only the currently valid update difference information 1704 (the figure that has not disappeared) and further communication. Time and network load can be reduced.

Finally, a method for searching for geographic information including history information at high speed will be described.
As described above, the geographic information storage method including time history information of this method has the advantages of (1) reducing the information storage area and (2) speeding up the search including time.
However, there are many search opportunities that do not include time, such as displaying the landscape of the area. In particular, in landscape display, high-speed search is essential because it is necessary to redraw the geographic information by changing the viewpoint in real time.
The four-dimensional geographic information including time is slow in search speed compared to the three-dimensional geographic information because past histories that are unnecessary for the landscape display are mixed in the figure and the constituent points.
In order to solve this problem, when a certain spatial region / time point is searched, the three-dimensional geographical information obtained by extracting only the figure existing at that time point is reconstructed from the original four-dimensional geographical information of the region.
The reconstructed three-dimensional geographic information is stored in a temporary area and can be used as an original alternative to a search that requires high speed, such as a landscape display, and that does not perform a time axis search.

As described above, the graphic data management method of the present invention can greatly reduce the data size of geographic information as compared with the conventional method.
In addition, it is possible to extract geographic information at an arbitrary time by adding time information indicating when each update target has changed when updating geographic information.

It is a schematic diagram which shows the characteristic of this invention. It is a block diagram of the geographic information processing system to which this invention is applied. It is a figure which shows the relationship of area | region, a figure, a composing point, and attribute information which comprises geographic information. It is a figure which shows the data structure of area | region, a figure, a composing point, and attribute information which comprises geographic information. It is a figure for demonstrating the method of the "generation / annihilation" time extension with respect to a figure. It is a figure for demonstrating the method of the "change" time extension with respect to a figure. It is a figure which shows the example of the "change" time extension of a figure. It is a figure for demonstrating the method of the "replacement" time extension with respect to a figure. It is a figure for demonstrating the method of the "generation / annihilation" time extension with respect to a composing point. It is a figure for demonstrating the method of the "change" time expansion with respect to a composing point. It is a figure for demonstrating the method of the "replacement" time extension with respect to an area | region. It is a figure for demonstrating the method of the "change" time extension with respect to an attribute. It is a figure for demonstrating the representation method of the figure which connects between areas in a XY coordinate direction. It is a figure for demonstrating the representation method of the figure which connects between areas in a Z coordinate direction. It is a figure for demonstrating the method of the area | region division in the case of dividing | segmenting an area | region for search speed improvement. It is a figure for demonstrating the method of area division | segmentation in the case of saving and storing the past history in a backup medium. It is a figure for demonstrating the method of the area division | segmentation in the case of updating geographic information between a client and a server. It is a figure for demonstrating the conventional method for performing time management.

Explanation of symbols

201 Display 202 Keyboard 203 Mouse 204 Printer 205 Scanner 206 Application Unit 207 DB Management Unit 208, 212 Computer 209, 211 Geographic Information DB
210 LAN

Claims (7)

A graphic data management method for managing each of a plurality of features such as houses and roads as graphic data, and registering and managing these graphic data in a database,
For each figure data of the feature, the generation time and disappearance time data of the feature are associated and held,
Accepts a graphic data transfer request from the client system that specifies the period from the first time to the second time ,
If the client system has already acquired graphic data at a time before the first time in the database , graphic data at a time after the first time acquired in response to the request A graphic data management method comprising: transferring graphic data of the feature including at least one of the generation time and the extinction time to the client system by the second time . 2. The graphic data management method according to claim 1, wherein the graphic data to be transferred is data representing a difference from the graphic data acquired by the generation or disappearance . The client system that has received the change difference of the graphic data adds the data representing the difference of the graphic data to the graphic data of the time before the first time and adds the information history of the feature up to the specified time. graphic data management method according to claim 2, wherein the building.   2. The graphic data management method according to claim 1, further comprising: associating and holding change time data for each graphic data of the feature. If the feature has disappeared, store the disappearance time as the disappearance time data,
2. The graphic data management method according to claim 1, wherein, when the feature exists, an identifier indicating that the disappearance time is undetermined is stored as the disappearance time data. Each of a plurality of features such as houses and roads as graphic data, a graphic data management system having a server that stores these graphic data in a database, and a database management unit that manages the database of the server,
The server
Corresponding data of the generation time and disappearance time of the feature for each graphic data of the feature is held in the database,
The database management unit
Accepts a graphic data transfer request from the client system that specifies the period from the first time to the second time,
If the client system has already acquired the graphic data of the previous time from the first time in the database, graphic data time after the first time in the database in response to the request And feature data including at least any one of the generation time and the annihilation time by the second time is extracted from the database of the server,
A graphic data management system, wherein the extracted graphic data is transferred to the client system. If the feature has disappeared, store the disappearance time as the disappearance time data,
7. The graphic data management system according to claim 6, wherein, when the feature exists, an identifier indicating that the disappearance time is undetermined is stored as the disappearance time data.

Images