KR102026888B1

KR102026888B1 - Method and apparatus for searching aerial image using dual mesh structure and file structure

Info

Publication number: KR102026888B1
Application number: KR1020120120981A
Authority: KR
Inventors: 한재성; 정혜영
Original assignee: 팅크웨어(주)
Priority date: 2012-10-30
Filing date: 2012-10-30
Publication date: 2019-09-30
Also published as: KR20140054910A

Abstract

Disclosed are an aerial image retrieval method and apparatus using a double layered structure, and a file structure thereof. An apparatus for retrieving an image of a target area from a file in which a map image is stored may include:-the file is a mesh in which the tile has a unit number with respect to a tile in which the map image is divided into unit sizes and stored; A tile search unit for searching for a target tile which is a tile corresponding to the target area; A leaf search unit for searching for a leaf where the target tile is located; And an image search unit for searching for an image corresponding to the target tile in a file in which the searched map leaf is located through the file structure.

Description

TECHNICAL AND APPARATUS FOR SEARCHING AERIAL IMAGE USING DUAL MESH STRUCTURE AND FILE STRUCTURE}

Embodiments of the present invention relate to a map image retrieval method and apparatus for quick retrieval and a file structure thereof.

Aerial images are widely used as the background of digital maps without any additional drawing. This is because the numerical map based on the live-action is advantageous for the user's map interpretation and location confirmation.

Korean Laid-Open Patent Publication No. 10-2006-0074601 (published on July 03, 2006) discloses a technique of using satellite photographs or aerial photographs as a background of a map image.

In the map service, the aerial view function is a service implemented to view the view from the sky 360 degrees in all directions. If the aerial view service is provided locally at the mobile terminal, the aerial image must be stored inside the terminal.

However, storing all tiled aerial images in one file is very large and impossible in terms of physical and technical aspects. In addition, the number of aerial images is also enormous given all serviceable levels. Therefore, storing all tiled aerial images into a single file may not be effective for searching for necessary images.

It is an object of the invention to design a file structure that can quickly retrieve aerial imagery within a given physical-technical environment.

It is possible to provide a double layered structure that makes searching for a desired map image faster and more efficient.

According to an embodiment of the present invention, an apparatus for retrieving an image of a target area from a file in which a map image is stored, the file may include: A tile search unit configured to have a file structure formed by an upper layer of the tile having a unit number of meshes, the tile search unit searching for a target tile corresponding to the target area; A leaf search unit for searching for a leaf where the target tile is located; And an image search unit for searching for an image corresponding to the target tile in a file in which the searched map leaf is located through the file structure.

According to an aspect, the tile search unit may calculate a tile index list intersecting the target area by using a resolution and coordinates according to the image level of the target area.

According to another aspect, the leaf search unit may calculate the leaf index of the target tile using the size of the leaf.

According to another aspect, the image search unit searches for the leaflet data corresponding to the leaflet index using leaflet offset information in the data file including the leaflet index, and then extracts tile offset information from the leaflet data. An image corresponding to the target tile may be searched for using the image.

According to an embodiment of the present invention, in the file structure in which the map image is stored, the file in which the map image is divided into unit sizes of the file in which the map image is divided into units of the tile is stored. It has a file structure composed of a higher layer, and when the map image is searched, the file structure corresponding to the search target may be accessed through the file structure, and then the image corresponding to the tile may be accessed in the map leaf.

According to one aspect, the file is composed of at least one data file, each of the data file is composed of a plurality of leaf data, and a leaf offset table indicating the location of each of the leaf data, each of the leaf data Tile data and a tile offset table indicating a position of each tile data.

According to another aspect, the file may further include a meta file, and the meta file may include a map index list included in each of the data files.

According to another aspect, each of the tile data may be composed of image data and state information indicating the state of the image data.

According to an embodiment of the present invention, a method of retrieving an image of a target area from a file in which a map image is stored comprises:-the file is divided into a unit size and the tile is stored in a tile. A mesh having a unit number has a file structure configured as an upper layer of the tile, and-searching for a target tile which is a tile corresponding to the target area; Searching for a leaf on which the target tile is located; And searching for an image corresponding to the target tile in the file in which the searched map leaf is located through the file structure.

According to an exemplary embodiment of the present invention, a map structure in which a map image is divided into unit sizes may be configured as an upper layer of tiles by configuring a map consisting of a unit number of tiles, thereby providing a faster and more efficient file structure for searching for a required image.

According to an exemplary embodiment of the present invention, a tile offset load for finding an image can be reduced by utilizing a double-leaflet structure of a method of finding a leaf in which a necessary image is located first and finding an image in a file in which the leaf is located.

According to an embodiment of the present invention, by using a double-leaflet structure, the entire map image can be divided and stored in a plurality of files, or an image included in one level can be divided and stored in a plurality of files. This is possible.

1 and 2 are diagrams for explaining a double leaf structure of a map according to an embodiment of the present invention.
3 and 4 illustrate a pile structure designed as a double lobe structure according to one embodiment of the present invention.
5 is a flowchart illustrating a map retrieval method using a double leaf structure according to an embodiment of the present invention.
6 and 7 are diagrams for describing a process of searching for a tile list with respect to a search area according to an embodiment of the present invention.
8 to 10 are diagrams for explaining a process of searching for image data in a map including a tile list according to one embodiment of the present invention.
11 is a block diagram illustrating an internal configuration of a map retrieval apparatus using a double leaf structure according to an embodiment of the present invention.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

The present embodiment provides a file structure for searching a map image more quickly. In the present specification, the 'map image' may refer to any type of map that is displayed as an image, such as a 2D map, a 3D map, an aerial photo.

The embodiments may be applied to a file system for storing a map image and a search system for searching for a map image. For example, the embodiments may be applied to a navigation terminal that provides a view mode using aerial photography.

1 is an exemplary diagram for explaining a double leaf structure of a map.

The map image stores the entire map in a grid of a certain size. Referring to FIG. 1, a tile 101 may mean an image unit included in one area when a map image is divided into a unit size that is a minimum block unit. In this case, the tile 101 has location information and may represent an index of an image of each region.

In addition, the mesh 103 may be an upper block unit with respect to the minimum block unit, and may include a tile number of units. In other words, the map leaf 103 refers to a set of tiles 101, and may mean an image unit included in one area when the entire map is divided into a grid larger than the minimum block unit.

The number of tiles 101 included in the map leaf 103 may be defined in advance, and the size of the map leaf 103 may be determined according to the total number of tiles for the entire map. In this case, the total number of tiles may be represented by (number of leaves x number of tiles per leaf). For example, if the number of maps is 1600 and the number of tiles per map is 1600, the total number of tiles for 12 levels is 256 million.

As shown in FIG. 2, when the number of leaflets is large (210), the size of the leaflets is reduced, resulting in a small memory usage but an increase in the number of input / output of files. On the contrary, when the number of the leaflets is small (220), the size of the leaflets is increased so that the memory usage increases, but the number of input / output of the file decreases. Therefore, the size of the leaf can be determined by considering the memory usage and the number of file input / output according to the system environment or condition.

When looking at the set of the map leaf 103 as a full map, in order to search for the required image in the entire map image, first find the map in which the image is located, and if only the image found in the map only to find the image in the map that is efficient Search is possible.

Referring back to FIG. 1, when the entire map is composed of one file 105, the tiles 101 are gathered together to form the leaf 103 and the leaves 103 are gathered to form the whole map as one file 105. It can be said that. On the other hand, when the whole map is divided into a plurality of files 105, the tiles 101 are gathered together to form the leaf 103, the leaves 103 are gathered to be one file 105, and the files 105 are all. It's a whole map.

As such, in view of the fact that the pile 105 is a collection of the leaflets 103, the leaflet 103 may mean a primary leaf and the pile 105 may mean a secondary leaflet, which may be referred to as a dual leaflet structure. .

In the present embodiment, the leaf structure described with reference to FIG. 1 may be applied to a file structure for storing a map image. In other words, in order to search for the necessary image in the entire map image, first find the map leaf 103 in which the image is located, find the file 105 in which the leaf map 103 is found, and then find the image in the file 105. If you can, you can search for images efficiently. Furthermore, if the map image is searched using the above double lobe structure, for example, the entire image of one level can be divided and stored in multiple files, thereby avoiding physical file size limitation.

3 shows a file configuration reflecting a double layered structure, which is an example of a file in which an aerial photo is stored.

One file may include a data file 310 and a meta file 320. At this time, the map image may be stored for each map level in the data file 310 (311), and the map image belonging to one level may also be divided and stored (312). In the present embodiment, each data file 312 may store a map image 314 defined as a tile in the map unit 313. In other words, the data file 310 has a file structure in which the map leaf 313 is composed of a higher layer of the tile 314 with respect to the tile 314 stored by dividing the map image into units of minimum blocks.

In more detail, referring to FIG. 4, the meta file 320 may include a system level number, a level-specific mapping table, an image level number, and level image information (first to Nth). In this case, the level-specific mapping table 321 may include a system level (first to Nth), an image level (first to Nth), and the like. In addition, the level image information 322 may include a resolution, a coordinate system (origin, basis), the number of tiles (horizontal and vertical) of a map leaf, the number of files, a list of files, and the like. Here, the file list 323 may include a data file name (first to Nth), a leaf index per data file (first to Nth), and the like.

What files are included in each data file when the file is generated may be recorded through the meta file 320. In other words, the file list 323 of the meta file 320 may include a map index list for each data file, thereby searching for which data file a particular map exists.

In particular, as shown in FIG. 4, each data file 312 may be composed of a plurality of leaf data (first to Nth) 313, and a leaf offset table 313a. In this case, offset information indicating a position in a corresponding file of each of the leaflet data may be maintained in the leaflet offset table 313a. Each leaf data 313 may include a plurality of tile data 314 and a tile offset table 314a. Here, the offset information indicating the position in the corresponding map of each tile data may be maintained in the tile offset table 314a. In addition, each tile data 314 may include image data corresponding to each tile and state information about the image data. For example, the image data may be composed of webp data, and the status information is an enum value of 1 byte. The image does not exist (0x0000), the image exists and is used (0x0001), and the image exists but is not used ( 0x0002), not the map area itself (0x0003), and others (0x0004).

In the present embodiment, a map consisting of a unit number of tiles for a tile in which a map image is divided into a unit size is first searched for a map where an image to be searched is located through a file structure composed of upper layers of tiles, and a file where the map is located is located. Once found, you can implement a way to search for images within that file.

If you do not use the lobe structure, which is the upper block of the tiles, excessive memory is required to load the tile offset for the entire image included in the file in order to retrieve the necessary image. For example, assuming that 4 bytes (int) are consumed to store one offset information, about 10 MB (= 2.5 million x 4 bytes) of memory is required for 2.5 million map images at 12 levels. However, when the layer structure is used as the upper layer of the tile, the amount of tile offset load for finding the required image can be reduced and the image search can be performed faster.

Hereinafter, a map search method using a double leaf structure will be described in detail. 5 is a flowchart illustrating a map retrieval method using a double leaf structure. In the map search method according to an exemplary embodiment, each step may be performed by the map search apparatus described with reference to FIG. 11.

step( S510 )

In operation S510, the map search apparatus may search for a tile corresponding to the target area to be searched. For example, when a map is displayed on the navigation terminal, a tile list intersecting with the current view area is found. Referring to FIG. 6, when the current view area corresponds to the reference sign 601, step S510 may be a process of searching for tiles 603 in which some or all areas overlap with the view area 601.

For example, the map search apparatus may find a required tile list by using an extent of the target area and a resolution and coordinates according to the image level of the target area. In this case, the area size may be calculated by itself in the DP engine such as the navigation terminal. In the case of tile coordinates, the area size may be calculated using a coordinate system used for image tiling.

As illustrated in FIG. 7, the map retrieval apparatus may use an image level of the current screen level 701 through a level mapping table 702 with respect to the current scene level 701. 703 may be recognized. The map search apparatus may find a tile resolution 707 corresponding to the image level 703 by using a level resolution table 704. In other words, the tile resolution 707 can be easily found in the table configuration shown in Table 1. In addition, the map search apparatus may find the tile coordinates 706 corresponding to the image level 703 by using a level coordinate table 705. Accordingly, the map search apparatus intersects the tile index list with the view region through the view region size 708 and the tile index calculation 709 using the tile coordinate 706 and the tile resolution 707 previously found. 710 can be obtained.

step( S520 )&step( S530 )

Referring back to FIG. 5, in operation S520, the map search apparatus may search for a map in which the tile list found in operation S510 is located. As described above, since the size of the leaf which is the upper layer of the tile is determined, the map search apparatus may calculate the mesh index of the tile list found in step S510 by using the size of the leaf. Referring to FIG. 8, step S520 may be a process of finding a map leaf 805 in which tiles 803 intersecting the view area are located.

In operation S530, the map search apparatus may search for a file in which the leaf found in operation S520 is located, and then search for an image corresponding to a tile to be searched for in the corresponding file.

9 shows a basic process of finding an image, and FIG. 10 shows a detailed process of steps S520 and S530.

Referring to FIG. 9, the map retrieval apparatus first loads the leaf offset table of the data file (S910) to determine where the leaf data to be found through the leaf offset table is located in the data file, and then loads the leaf data from the corresponding position. (S920). Subsequently, the map retrieval apparatus loads the tile offset table of the loaded leaf data (S930), determines where the tile data to be found through the tile offset table is located in the leaf data, and loads the tile data at the corresponding position (S940). Finally, the map search apparatus loads image data from the loaded tile data (S950), thereby completing the image search process in the double-leaf structure.

In more detail, as illustrated in FIG. 10, the map search apparatus receives the tile index list 1001 found in step S510, and an image is displayed for each tile index 1002 included in the tile index list 1001. It may be checked whether the state 1003 is already loaded. The map retrieval apparatus may find the leaf index 1006 where the corresponding tiles are located through the leaf index calculation 1005 using the leaf size 1004 for the tile list on which the image is not loaded. In this case, the map retrieval apparatus may determine which data file 1007 includes the leaf data corresponding to the leaf index 1006 previously found through the leaf index list included in the meta file. As described above, the map retrieval apparatus loads the leaflet offset table indicating the position of the leaflet data in the data file 1007 as described above, and then, through the leaflet offset table, the data file for the leaflet data corresponding to the leaflet index 1006. The location within 1007 can be confirmed. Subsequently, the map retrieval apparatus loads the tile offset table indicating the position of each tile data in the corresponding map data (1008), and then uses the tile offset table 1009 to map the tile data corresponding to the tile index 1002 in the map data. The location 1010 may be checked to load the desired tile data 1011.

step( S540 )

Referring back to FIG. 5, in operation S540, the map search apparatus may decode an image found through the above process and make it a texture. For example, image data included in tile data may be stored in webp format for efficiency of file capacity, which can be easily decoded in real time. In this case, the map search apparatus may temporarily store the decoded image for future reuse. In addition, the map search apparatus finds four vertex coordinates of the image found through the double-leaflet structure and draws them on the screen. In this case, four vertices of the image may mean four vertices of the tile.

Therefore, in the present exemplary embodiment, the necessary image may be quickly found by using a file structure in which a map consisting of a unit number of tiles is formed as a higher layer of the tile with respect to a tile in which the map image is divided into unit sizes.

11 is a block diagram showing an internal configuration of a map retrieval apparatus using a double leaf structure. As illustrated in FIG. 11, the map search apparatus 1100 according to an embodiment may include a tile search unit 1110, a map search unit 1120, an image search unit 1130, and a decoding unit 1140. Can be configured.

The map search apparatus 1100 searches for a map image stored in a file system having the structure described with reference to FIGS. 3 and 4.

The tile search unit 1110 may serve to search for a target tile that is a tile corresponding to the target area. For example, the tile search unit 1110 corresponds to the target tile through a method of calculating a tile index intersecting with the current view area by using the size of the current view area that is the target area, tile resolution and tile coordinates corresponding to the image level. You can get a list of tile indexes.

The leaf search unit 1120 searches for a leaf where the target tile is located. For example, the leaf search unit 1120 may calculate the leaf index of the tile index list using the size of the leaf.

The image search unit 1130 searches for a file in which the corresponding leaf is located with respect to the leaf in which the target tile is located, and then searches for an image corresponding to the target tile in the file. As an example, the image search unit 1130 may search for the map leaf index list in the meta file stored in the file to find a data file in which data data corresponding to the leaf index found by the leaf search unit 1120 is stored. Subsequently, the image search unit 1130 checks the position in the data file of the leaf data using the leaf offset information stored in the data file, and checks the position in the leaf data of the tile data using the tile offset information stored in the leaf data. By doing so, you can find the tile data you want to find.

The decoding unit 1140 decodes an image corresponding to the target tile and serves as a texture. For example, the decoding unit 1140 may create a texture by decoding an image stored in a webp format. In this case, four vertices of a tile are applied to the decoded image as four vertices constituting the image and easily drawn on the screen. can do.

As described above, according to an exemplary embodiment of the present invention, a map structure in which a map image is divided into unit sizes may be configured as an upper layer of tiles by constructing a map consisting of a unit number of tiles to provide a faster and more efficient file structure for searching for a necessary image. Can be. In addition, according to an embodiment of the present invention, the amount of tile offset load for finding an image may be reduced by utilizing a double-leaflet structure of a method of finding a leaf in which a necessary image is located first and finding an image in a file in which the leaf is located. In addition, according to an embodiment of the present invention, by using a double-leaflet structure, the entire map image may be divided and stored in a plurality of files, or an image included in one level may be divided and stored in a plurality of files. Image search is possible.

Methods according to an embodiment of the present invention may be implemented in program instruction form that can be executed by various computer systems and recorded in a computer readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. Program instructions recorded on the media may be those specially designed and constructed for the purposes of the present invention, or they may be of the kind well-known and available to those having skill in the computer software arts. In addition, the above-described file system can be recorded in a computer-readable recording medium.

As described above, the present invention has been described by way of limited embodiments and drawings, but the present invention is not limited to the above embodiments, and those skilled in the art to which the present invention pertains various modifications and variations from such descriptions. This is possible.

Therefore, the scope of the present invention should not be limited to the described embodiments, but should be determined not only by the claims below but also by the equivalents of the claims.

1100: map search device
1110: tile searcher
1120: leaf search
1130: image search unit

Claims

In the device for retrieving the image of the target area in the file (file) stored the map image,
The file has a file structure in which a mesh consisting of a unit number of tiles is formed as an upper layer of the tile with respect to a tile in which the map image is divided into unit sizes and stored.
A tile searching unit searching for a target tile corresponding to the target area;
A leaf search unit for searching for a leaf where the target tile is located; And
An image search unit for searching for an image corresponding to the target tile in a file in which the searched leaf is located through the file structure
Map search device comprising a.

The method of claim 1,
The tile search unit,
Calculating a tile index list that intersects the target area using a resolution and coordinates according to the image level of the target area;
Map search device characterized in that.

The method of claim 1,
The leaf search unit,
Calculating the leaf index for the target tile using the size of the leaf
Map search device characterized in that.

The method of claim 3,
The image search unit,
After searching for the leaf data corresponding to the leaf index in the data file including the leaf index, the image corresponding to the target tile is searched using the tile offset information in the leaf data. To do
Map search device characterized in that.

In the file structure where the map image is stored,
The file where the map image is stored,
For a tile stored by dividing the map image into unit sizes, a map structure having a tile number of units has a file structure in which an upper layer of the tile is formed.
When searching for an image of a target region in a file in which the map image is stored, the target tile, which is a tile corresponding to the target region, is searched for, and then, the searched map where the target tile is located to locate the searched leaf through the file structure. Searching for an image corresponding to the target tile in a file
File structure characterized in that.

The method of claim 5,
The file consists of at least one data file,
Each of the data files is composed of a plurality of leaf data, and a leaf offset table indicating a position of each of the leaf data.
Each of the sheet data comprises a plurality of tile data and a tile offset table indicating a position of each of the tile data.
File structure characterized in that.

The method of claim 6,
The file further includes a meta file,
The meta file includes a list of leaf indexes included in each of the data files.
File structure characterized in that.

The method of claim 6,
Each of the tile data is composed of image data and state information indicating a state of the image data.
File structure characterized in that.

In the method for retrieving the image of the target area from the file where the map image is stored,
The file has a file structure in which a map consisting of the number of units of a tile is configured as a higher layer of the tile with respect to a tile in which the map image is divided into unit sizes and stored.
The method for searching for an image of the target area may include
Searching for a target tile that is a tile corresponding to the target area;
Searching for a leaf on which the target tile is located; And
Searching for an image corresponding to the target tile in a file in which the searched map is located through the file structure
Map search method comprising a.

The method of claim 9,
Searching for the target tile,
Calculating a tile index list that intersects the target area using a resolution and coordinates according to the image level of the target area;
Map search method characterized in that.

The method of claim 9,
Searching for the lobe where the target tile is located,
Calculating the leaf index for the target tile using the size of the leaf
Map search method characterized in that.

The method of claim 11,
Searching for an image corresponding to the target tile,
Searching for leaf data corresponding to the leaf index by using leaf offset information in a data file including the leaf index; And
Searching for an image corresponding to the target tile using tile offset information from the map data
Map search method that includes

The method of claim 9,
The file consists of at least one data file,
Each of the data files is composed of a plurality of leaf data and a leaf offset table indicating a position of each of the leaf data.
Each of the sheet data comprises a plurality of tile data and a tile offset table indicating a position of each of the tile data.
Map search method characterized in that.

The method of claim 13,
The file further includes a meta file,
The meta file includes a list of leaf indexes included in each of the data files.
Map search method characterized in that.

A computer-readable recording medium in which a program for executing the method of any one of claims 9 to 14 is recorded.