JPS61170868A - Demarcating method of unit picture in overall picture - Google Patents

Abstract

PURPOSE:To prevent drop-out of information by defining four mesh areas which are adjacent to one another vertically and horizontally as one unit picture to provide an overlapping area between unit pictures adjacent to each other. CONSTITUTION:With respect to each area whose square area is determined by a mesh Me, four mesh areas adjacent to one another vertically and horizontally form one unit picture. Consequently, a maximum of 9 unit pictures can be taken if meshes Me demarcated with 5 X-axis direction dividing lines and 5 Y-axis direction dividing lines are considered. In the case, picture information Qa which exists across the boundary between unit pictures (b) and (c) is placed in a complete form on a unit picture (e) or (h), and picture information Qb which exists across the boundary between unit pictures (c) and (d) is placed similarly on a unit picture (g), and picture information Qc existing across the boundary between unit pictures (a) and (c) is placed in the center of a unit picture (f) in a complete form.

Description

[Detailed Description of the Invention] <Industrial Application Field> The present invention provides an overview of the entire image, which is necessary when storing an entire image having a predetermined area, such as various maps, in an image file system as a set of unit images. The present invention relates to a method for defining unit images, and particularly to a method for defining unit images that is improved so that adjacent unit images have regions that partially overlap each other.

<Prior art> In recent years, image file systems have been developed that have non-volatile external memory with large capacity memories such as optical disks. , city name, town T name, block number 1
By inputting identifiable data such as street addresses, landmarks, building names, building ownership deeds, etc., the software displays a partial map on the display or hard copy that includes the designated area or location according to the data. It has become widely used.

Naturally, in such a system, it is necessary to input images to the image file system in units of each partial map. Therefore, first, each partial map is divided into appropriate areas from the entire map as an honor image. must be created extractively.

However, the method of extracting or defining unit images conventionally used for this purpose was extremely simple and common, as shown in FIG.

That is, the entire image Gw is placed on a suitable two-dimensional XY plane, and the image Gw is divided by a mesh M having n and m dividing lines in the X-axis direction and the Y-axis direction, respectively, and The area included in each mesh Me whose four sides are defined by two dividing lines and two adjacent dividing lines in the Y direction was defined as each unit image.

In Figure 3, among these unit images, four unit images a
, b, c, and d are shown surrounded by thick lines, and the fourth
They are shown enlarged in the figure.

In FIG. 4, only two diagonally located unit images a and d are shaded for ease of viewing.
For convenience, in comparison with the method of the present invention described later, the numbers of the x and Y division lines are treated as the respective X and Y coordinate values, and the coordinate values P (x, y) are assigned to each intersection of the mesh M. It shows. In particular, the coordinate values of the coordinate points that are the common corner points among the four corners of the four unit images a, b, c, and d shown in the figure! ”' ＋
Shown as Y-J.

(Problems to be Solved by the Invention) When using the conventional unit image definition power as described above, the problem is the existence of image information Q spanning two or more adjacent unit images. be.

For example, in the case shown in the figure, if we consider the case where the image information Q exists across the boundary between the unit image and the unit image C, by inputting some specified data,
image file system or 5o) image information. 8□2゜Information□□Yo - However, as long as the conventional unit image definition method as described above is used, each unit image is originally inputted individually into the image file system. No matter what kind of reading software is used, unless two unit images (a) and (c) are simultaneously read out on one screen, part of the desired image information Q will always be missing.

In other words, in the conventional method described above, in which the area whose area is defined by each mesh of the meshes used to segment the entire image is directly defined as each unit image, the side part of each unit image is When the image information Q located across the two sides is read out, there will always be a loss of information.

In view of this point, the present invention has image information located across adjacent unit images, and therefore, among those unit images,
No matter which one is read out, even if there is a risk of missing information in the image information, the target image information will remain almost the same in another unit image without any missing information. The purpose of this invention is to provide a rational and unambiguous method for defining unit images that can be used to obtain unit images that are adjacent and partially overlap each other, as can be expressed in a simple manner.

<Means for Solving the Problems> In order to achieve the above object, the present invention provides a method for selectively defining a unit image having a unit area smaller than the predetermined area in a whole image having a predetermined area. Segment the above-mentioned entire image with a mesh having a plurality of dividing lines in the X-axis direction and the Y-axis direction;
In each region group whose rectangular area is defined by each of a plurality of meshes whose four sides are defined by two dividing lines adjacent in the axial direction and two dividing lines adjacent in the Y-axis direction, the upper and lower A total of four adjacent mesh areas, two on the left and right, and two on the left and right, are combined into one unit image; thus, adjacent unit images have areas that partially overlap with each other. A method for defining a unit image in an entire image is provided.

(Function) As is apparent from the above summary structure, in the present invention, one area itself defined by the mesh used for segmenting the entire image as in the past is treated as a unit image. In each region group whose rectangular area is defined in each mesh, there are two upper and lower regions, which are not demarcated.
Two on the left and right.

A total of four adjacent mesh areas are combined into one unit image.

In this way, one unit image is defined by four mesh areas, which means that two of the four mesh areas that define the unit image, two arranged on the left and right, or two arranged above and below, Alternatively, this means that each one may be used as two or one mesh area of other adjacent unit images, respectively.

Therefore, for example, even if there is image information that spans two adjacent unit images, the image information will always be present in one of the other unit images that partially overlap with these unit images. be damaged in any way,
It will arrive in perfect condition.

Therefore, even if the basic principle of reading only the middle unit image is followed as in the past, image information that has a portion protruding from the edge boundary is not searched in any other unit image. However, according to the present invention, image information that is only partially contained in a unit image is always read out completely in one unit image. Since the unit image is contained in the form, even under the above-mentioned limitation of reading only one image, it is sufficient to read out the unit image contained in this complete form.

Of course, if the number of dividing lines in the X-axis direction and the Y-axis direction constituting the mesh is the same as in the conventional method, when the same overall image is divided at the same scale, the present invention The unit area of each unit image is four times that of the conventional one. . ,,wow, 3 months 6
．． □2□Yu. If J-fineness is required, simply increasing the number of 0 dividing lines by doubling the number of dividing lines in the X-axis direction and Y-axis direction that make up the mesh compared to the conventional method. There is no technical difficulty in itself.

<Embodiment> FIG. 1 shows the principle or basic embodiment of the present invention. As explained in conjunction with FIG. 3, this FIG. This is an enlarged view of a portion of the figure.

In the present invention, as described above, each region group whose rectangular area is defined by each of the (m-1) x (21-1) meshes Me, . A total of four adjacent mesh areas, two in the middle, two on the top and bottom, and two on the left and right, are combined to form one unit image.

Therefore, as shown in FIG. 1, when considering a total of 16 meshes Me, . . . defined by five X-axis direction division lines and five Y-axis direction division lines.

Up to nine unit images can be taken.

That is, first, four unit images a, b, c, and d with no overlapping parts are obtained, and in addition, each of these four unit images a, b, c, and d shares one mesh area. In addition to obtaining the unit image e shown with a diagonal line downward to the left in the center in the drawing, the single unit images a, b, c, and d are located vertically, horizontally, and vertically relative to each other. A total of four unit images f 9g * h * 1, with two mesh-eye areas taken from adjacent ones.
(Virtual line in the figure) can also be obtained.

Therefore, for example, as schematically shown in the figure, when considering only individual unit images a, b, c, and d, image information Qa, Qb,
Even if there is Qc, such image information will be located approximately in the C central area in any of the other unit images that partially overlap with these unit images.

For example, image information Qa that straddles the boundaries between unit images S and C will appear in its entirety in unit images e or h, and similarly, image information Qb that straddles between unit images c and d will appear in unit image g. In addition, the image information Qc extending between unit images a and c is located in the unit image f in a complete form and at the center of the unit image area.

Therefore, these unit images a-
If you input and memorize f, and associate the optimal unit image with each image information Qa-Qc with an appropriate designation code or data, you can use it for any desired image information in the entire image 0w. However, even if the unit image is not lost, it is possible to read out the unit image without fail.

Book L understood the invention from the perspective of the Eye of Metush,
It is also possible to understand it by expanding it to a general coordinate system.

That is, if we consider the numbers of the dividing lines in the ,j);
l≦i≦n-1, 1≦j≦m-1, and the coordinate values of the four corners are P(i-1, j◆1) and P(i-1, j-1), respectively.
).

It can be said that the present invention defines an area defined by five coordinate values p(i+t, J++) +P('÷1.j-1) as one unit image.

Taking correspondence with respect to such a coordinate system, in the conventional unit image definition method shown in FIG. 4, for example, a unit image a has four coordinate values P(i, j) and P(i-1, j+1).

As can be seen from the fact that it is defined by P(i-1,j) and P(i,j÷1), each unit image is It merely defines the

In any case, if each unit image is defined and considered using a coordinate system in this way, each unit image can be managed using the center coordinate values of each intermediate image depending on the purpose or use. , I can also do various convenient things.

Of course, even after the entire image Gy is divided into meshes M, the present invention only needs to be applied to the necessary areas, and the present invention is applied only to the areas where there is no image information that spans between the mesh areas. It is not necessary to apply the present invention,
In other words, even with respect to one coordinate point, it is not necessarily necessary to define all unit images having all the coordinate points P(i, j) as the central coordinate point.

FIG. 2 shows an example in which the present invention is applied to a residential map in order to explain the effectiveness of the present invention more clearly.

FIG. 2(A) shows a segmented partial image enlarged from the entire image, that is, a partial map. For ease of explanation, the dividing lines are numbered 1 to 4 in both the X-axis direction and the Y-axis direction.
Assign numerical values up to the number and consider according to the coordinate system.

In the illustrated case, there are nine mesh areas, but according to the idea of the present invention, four unit images can be extracted from these areas. These are shown six times from FIG. 2 (CB) to FIG. 2 (E), and each unit image or unit map section is indicated by its center coordinate value.

That is, Fig. 2 (B) is a unit image with center coordinate value P (2, 2), Fig. 2 (C) is a unit image with center coordinate value P (3, 2), and Fig. 2 (D) is a unit image with center coordinate value P (3, 2). Unit image of coordinate value P (2, 3),
FIG. 2(E) is a unit image with center coordinate values P(3, 3).

Looking at FIG. 2(A), it can be seen that there are quite a number of houses, etc. that straddle each mesh area.

However, with the conventional unit image definition method, houses separated by each division line cannot be completely represented no matter which mesh area is read out. If you look at the "painting materials" building shown by the diagonal circle, or the building named by Tadae Daisumi,
These are the coordinate values of the four corners P(2,2) and P(2,1)
, P(3,1), P(3,2), and the coordinate values of the four corners to the right of it, P(3,
2) , PC3,1), P(4,1) , PC4,2)
Since it is divided by a conventional unit image defined only by , the target building cannot be read out in its complete form no matter which of these unit images is read out.

On the other hand, when searching among the unit images defined by applying the method of the present invention, it is possible to find the above-mentioned "four-phase" building or "Daikaku Keizon".
It can be seen that several buildings are shown in complete form in the central area of Figure 2 (C) without being divided into parts. If you conduct a similar study on buildings on other division lines, you will understand that they are always included in some unit image in their complete form.

In general, in the case of a residential map, first cut out the same medium-sized maps a, b, c, and d as shown in Figure 3 from the overall map as shown in Figure 3, and then This is then further segmented using a mesh to define a more detailed unit image in accordance with the present invention. Therefore, in such a case, it is sufficient to first consider each of the cut out medium-sized maps as the entire image as referred to in the present invention, and the gist and structure of the present invention can be applied as is.

<Effects of the Invention> As described above, according to the present invention, when managing an entire image as a set of unit images using an image file system, it is possible to ensure that the entire image is entered in a complete form no matter which unit image is searched for. The existence of missing image information can be fundamentally avoided, and the desired image information can be obtained in a complete form in any unit image.

In addition, since the meshes of the entire image are divided, two meshes each on the top, bottom, left and right, and a total of four adjacent meshes define one unit image, multiple unit images that partially overlap can be defined. However, it is possible to allocate it in an extremely rational manner, and there is no waste.

Therefore, for example, an image file such as a residential map etc.
Managed by the system, municipality name, town name, block number 1
By inputting identifiable data such as street addresses, landmarks, building names, building owner names, etc., a partial map containing the area or location specified according to the data is displayed on a display or hard copy. When the present invention is used for an application, image information corresponding to the input data, such as a house, appears only in a part of the unit image read out according to the data, unlike in the conventional case. Lever, t b'
* (7) 1″m 4!r t* CIn, t!
/f <': ,! : b<v! ! ,'! z
#1 As long as the relationship between the data and the corresponding unit image is correctly determined at the initial input stage to the image file system, a map of the vicinity with the target house on it will always appear in the center of the screen. can. still.

The center of the screen, as is usually said in this type of technology, means an easily recognizable area excluding the periphery.

[Brief explanation of drawings]

FIG. 1 is an explanatory diagram of the principle or basic embodiment of the present invention,
FIG. 2 is an explanatory diagram showing the specific effectiveness of the present invention, FIG. 3 is an explanatory diagram of segmentation of an entire image, and FIG. 4 is an explanatory diagram of a conventional unit image definition method. In the figure, Gw is a mesh of the entire image 1M, Me is a mesh, a-f are unit images, and Q, Qa, Qb, and Qc are image information. 11I1 ,,Gw↑Imoxibustion〉Comparison Station-Kai Xi

Claims (1)

[Scope of Claims] A method for selectively defining unit images having a unit area smaller than the predetermined area in a whole image having a predetermined area; Segmented by a mesh having a plurality of dividing lines; a plurality of meshes whose four sides are defined by two dividing lines adjacent to each other in the X-axis direction and two dividing lines adjacent to the Y-axis direction of the mesh; In each region group whose rectangular area is determined, a total of four adjacent mesh regions, two on the top and bottom and two on the left and right, are combined into one unit image; thus, unit images that are adjacent to each other. A method for defining unit images in a whole image, characterized in that the images have regions that partially overlap each other.