JPH03184170A - Document retrieval system - Google Patents

Abstract

PURPOSE:To retrieve a document in a retrieval screen by finding areas of a graph, a table, and an image when a document is inputted, generating and storing an index file according to the merging point for a previously divided area, and using the file. CONSTITUTION:When a document is generated by a document generation part 4, the area 33 of the graph is defined in a document 35 of characters 32 by an input processing part 1 such as a mouse for the figure 34. At this time, the graph 36, image 37, and table 38 are identified and specified and and a processing part 9 which calculates the area of the defined area and the area ratio from vertex coordinates found by an area definition processing part 8 according to the input position data. An area number arithmetic part 10 searches for the merging point for the divided area with the predetermined number and inputs the identification number of the graph, etc., to an input area 41 and the area ratio to an area ratio 21 on the retrieval screen 39 by using the index file generated with the number and various pieces of information. Consequently, the discrimination among the graph, table, and image and its position are only stored to easily perform the retrieval, thereby improving the operability.

Description

[Detailed description of the invention] [Industrial application field] The present invention includes two tables of figures created by one document creation machine.

This invention relates to a method for searching documents containing images.

[Conventional technology]

As described in Japanese Patent Laid-Open No. 60-160475, conventional devices have a method of dividing a document containing two images of figures and tables into areas containing homogeneous data, and searching for the document by inputting the structural information of the document from the keyboard. .

[Problem to be solved by the invention]

The above-mentioned conventional technology divides the data into regions where data of the same quality exists, and searches using the type, size, location, etc. of the data as keywords, and it is necessary to input these data from a keyboard at the time of the search. The present invention makes it easy to search for where a type such as a figure 9 table 2 image is located by simply specifying the location and data type at the time of search by setting divided areas in advance.
The purpose is to provide an easy-to-use search method.

[Means to solve the problem]

To achieve the above purpose, shapes when inputting documents.

The area of the Table 2 image is determined, an index file is created and stored based on the merge points with the previously divided areas, and this index file is used to enable document searches on the search screen.

[Effect]

By specifying the graphic 2 table 2 image and defining the area, the merging point I with the area divided in advance is determined and an index file is automatically generated. As a result, it is possible to search for a target document simply by inputting the specific position of the figure 2 table 2 image in the document page area.

〔Example〕

An embodiment of the present invention will be described below with reference to FIGS. 1 to 8.

In FIG. 1, there is an input processing section 1 comprised of a keyboard, mouse, etc. for inputting one document or inputting data. The data input here is controlled by the input control section 2. Input data is controlled by a data control unit 3 in order to execute processing according to the purpose. For example, when a document is created by performing document processing 2 editing, etc., data control is performed so that the processing is performed in the document creation section 4. These document data pass through a data control section, are controlled by a data output control section 5, and are then displayed on a data display processing section 6 such as a CRT.

When generating one figure, two tables, two images, or calling other data from the document storage unit 7 while creating a document, the document and data input processing unit 1 defines an area. A known method for defining the area is used. With this definition input, the input control unit 2. The vertex position data of the region is input to the region definition processing section 8 via the data control section 3 .

The area definition processing unit 8 calculates the vertex coordinates of the area based on this input. The determined vertex coordinates pass through the data control section 3, and then the area and area ratio calculation processing section 9 calculates the area ratio as a proportion of the document page area. In the area number calculation unit 10, in order to simplify the position specification of document page areas at the time of search,
Determine where it belongs in a predetermined and numbered divided area. The index file creation unit 11 creates a table consisting of this number, the identification number of the figure 9 table 1 image, the 1 area ratio, and the document number which is the key when searching for a document.
and stores it in the index storage unit 12.

On the other hand, one created document is also stored together with a key file in the document storage section 7 using the document number as a key.

The document stored in the document storage section 7 is stored in the data display processing section 6
Document the search data according to the displayed items.

After the data is input in the data input processing section 1 and controlled by the input control section 2, the search processing section 13 searches the data in the indexed storage section 12. The search uses a publicly known method. As for the search results, the search processing section calculates the number of target documents retrieved from the one-document storage section 7 using the document number as a key, and displays it on the data display processing section 6. The obtained document is searched and displayed based on the document number from the key file in the document storage section 7.

FIG. 2 shows the flow up to index file creation.

When a document is created 14 in the document creation section 4, as shown in FIG.

Perform step 5 on the table area definition. As shown in FIG. 7, the area definition 15 is performed by using the input processing unit 1 such as a mouse to define an area 33 for a graphic 34 in a character 32 in a document 35 displayed on a CRT 31.

At this time, the identification designation of the figure 362, image 372, and table 38 is also performed. Next, based on the input position data, the area and area ratio of the area defined from the vertex coordinates determined by the area definition processing section 8 are calculated 16.

On the other hand, in order to search for a merge point with the numbered divided areas 25 as shown in FIG. 5 using the vertex coordinates, the area number calculation section 10 performs calculations to obtain the area number 17. Any number of divisions is possible at this time.

Regarding the index file created by the process 18 in FIG. 2, taking FIG. 7 as an example, if the position of the divided area is 4 and it is identified and input as a figure, the table at that time will contain the table shown in FIG. 3. As shown, the document number ``;i20'' and the calculated area ratio 21 are input.
There are as many regions as the number of divisions of the divided region 25, and the number 1 of the figure is manually entered at the fixed coordinate of 4 as shown in FIG. - If there are a number of Figure 2 Table 2 images in the document, the processes 15 and 19 in Figure 2 will be repeated.

FIG. 6 shows the flow of the search. In the search screen output 26, a document search screen 39 as shown in FIG. 8 is output. Here, the identification number of the figure 9 image 2 table is input into the input area 41 in the divided area.

In addition, if necessary, the area ratio 42 is input. Search processing 2 for data in the index storage unit 12 based on this input data
Search by 8. In addition, when 5 area ratios are input, in order to absorb errors for the index area ratio 21, the area ratio value is given a width of plus or minus 10%, and the range search process 28 is changed to the area and area ratio calculation process. Part 9 will be held.

Based on these, the output 29 of the corresponding number of documents is output as the target number 43 in FIG. Position of input area or 1
Ambiguous search is also possible by changing the area ratio.

Further, the present invention can read 2 figures 2 images 2 tables and their positions. It is also possible to apply it to the text input position.

〔Effect of the invention〕

According to the present invention, even if you do not understand the contents of a document, you can easily search by remembering the distinction between 1 figure, 9 tables, and 2 images included in the document and their positions, and the operability of the search is improved. improves.

[Brief explanation of drawings]

Fig. 1 is an overall configuration diagram of an embodiment of the present invention, Fig. 2 is a flowchart of index file creation, Fig. 3 is a table configuration diagram of an index file, and Fig. 4 is a figure 2 image 2 table and their respective identification power distributions. Figure 5 is a diagram showing the division area ffi when the original paper is divided into six parts, Figure S3-Figure 6 is a diagram showing the flow during search, Figure 7 is a diagram showing the area definition display screen,
FIG. 8 is a diagram showing a search screen when searching for documents. 11... Index file creation section. 39...Document search screen. Hanaotsu No. 2 Kuni No. 2

Claims (1)

[Claims] 1. An input processing unit for inputting documents and data;
A data display processing unit that outputs documents, search screens, search results, etc., an input control unit that controls input data, a document creation unit that processes input data into documents, and a document storage that stores created documents. an area definition processing section that defines a figure, table, or image region and calculates the vertex coordinates of the region; an area ratio calculation processing section that calculates the ratio between the area of the region and the page area of the document from the determined vertex coordinates; An area number calculation unit that calculates which divided area it belongs to based on the coordinates, an index file creation unit that includes diagram identification input, area area ratio and area number data, an index file storage unit that stores index data, and index data. a search processing unit that searches for documents, a data control unit that controls input/output data, a data output control unit that generates data for outputting documents and data, and a data display for displaying data and documents. A document search method characterized in that a document processing device including a processing section searches for a target document by inputting the positions of figures, tables, and images.