JP2002342343A - Document managing system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a document managing system, by which an inputted document image can be automatically divided by applying a suitable attribute even when that image is not a routine form. SOLUTION: When a document image is inputted by an image input part 1, the document image is divided into sentence area, table area, ruled line or drawing area by an area information extracting part 2. A layout information extracting part 22 extracts information on a layout from the document image divided into areas by area identifying processing on the basis of a layout database 23. After the layout information is extracted, character recognition is executed to an area at a specified position by a character recognizing part 24. Simultaneously with character recognition, bar code recognition is executed by a bar code recognizing part 26. Information read by such character recognition or bar code recognition is added to the document image as an attribute and the document image is saved in a document image storage part 4. By using such an attribute, the document image can be retrieved.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a document management system for assigning and managing one or more attributes to an input document image.

[0002]

2. Description of the Related Art In general, in a document image management system such as an electronic document filing apparatus, a method of assigning and managing one or more attributes to an input document is often used. Conventionally, adding attributes to document images
Information that can be easily obtained by the document management system, such as the date and time when the document image was registered and the registrant who registered the image document, are used. In addition to the information that can be easily obtained by such a document management system, the addition of an attribute to a document image includes, for example, an operator directly inputting a character string or selecting an attribute to be added from a previously registered attribute list. In general, it is often performed manually. There is also a method of sorting documents by type before inputting them, and inputting them as document images collectively for each type. As a method of inputting and managing such a document image, Japanese Patent Application Laid-Open No. H10-240958 discloses a method in which a fixed form is identified, information included in the document is read, and the read information is associated with the document. A management information extracting apparatus and method for extracting management information from an image that can automatically extract information from a document by managing the information and reduce the work of an operator are described.

[0003]

However, a general manual method such as input of a character string by an operator or selection of an attribute to be added, documents are sorted by type before inputting, and documents are collectively collected. In the method of inputting as an image, when the type and amount of the document image increase, the load on the operator increases, which is inefficient. Also, in the invention described in Japanese Patent Application Laid-Open No. 10-240958, since the attribute to be added to the document is obtained by reading information described on the document, the attribute actually required is described in the corresponding document. It is not always the case, and accurate attributes may not be extracted in some cases. In such a case, it is necessary to manually add desired attributes after inputting the document. In order to reduce the operator's work at that time, attributes to be given to each fixed form of the input document are defined, and attributes given to the document image by identifying which form the input document image is. Although a method of determining the document image has been proposed, the document image input to the document image management system is not always a fixed form.

Accordingly, an object of the present invention is to provide an attribute by using area information of an input document image, thereby automatically assigning an appropriate attribute to the input document image even if the input document image is not a fixed form. It is another object of the present invention to provide a document management system capable of classifying a document image input later by an operator by adding attributes.

[0005]

According to the first aspect of the present invention, a document image acquiring means for acquiring a document image, and a document area, a character area, a table area, and a drawing area from the document image acquired by the document image acquiring means. Area information extracting means for extracting area information, such as, area information extracted by the area information extracting means, an attribute adding means for adding the area information as an attribute to the document image, and a document image to which the attribute is added by the attribute adding means The above object is achieved by providing document image storage means for classifying and storing the same for each attribute.

According to a second aspect of the present invention, in the first aspect of the present invention, the area information extracting means also determines the type and number of each area such as a document area, a table area, and a drawing area extracted as the area information. The attribute adding means extracts and adds the type and number of each area as attributes to achieve the above object. In the invention according to claim 3, claim 1
Alternatively, in the invention according to claim 2, the area information extracting unit extracts, as layout information, a position on a sheet of each area extracted as the area information, and the attribute adding unit adds the layout information as an attribute. By doing so, the above object is achieved. According to a fourth aspect of the present invention, in any one of the first, second, and third aspects of the present invention, the area information extracting means includes the area information extracting means included in a predetermined area of the area information. The character is extracted as character information, and the attribute adding means achieves the object by adding the character information as an attribute. In the invention described in claim 5, in the invention described in any one of claims 1, 2, 3, and 4, the area information extracting means adds a barcode to the document image. In this case, the barcode is extracted as barcode information, and the attribute adding means achieves the object by adding the barcode information as an attribute.

According to the present invention, a document image acquiring means for acquiring a document image and area information such as a document area, a character area, a table area and a drawing area are extracted from the document image acquired by the document image acquiring means. Area information extraction means, two or more pieces of the area information are combined, a document image classification means for classifying the type of the document image for each combination of the area information, and the area information extracted by the area information extraction means is attributed. Attribute adding means for adding to the document image as, a document image storing means for classifying and storing the document images to which the attribute is added by the attribute adding means based on the classification by the document image classifying means, With the provision of the above, the above object is achieved.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below in detail with reference to FIGS. FIG. 1 is a block diagram illustrating a schematic configuration of the document management system according to the first embodiment. As shown in FIG. 1, the document management system includes an image input unit 1 for inputting a document image using an image input device such as a scanner, an area information extraction unit 2 for extracting area information from the input document image, A document image management unit 3 for adding area information to a document image as an attribute and a document image storage unit 4 for storing an input document image are provided.

FIG. 2 is a diagram showing the area information extraction unit 2. The area information extracting unit 2 divides the document image into one or more areas such as sentences, figures, and tables according to the area information to be extracted as shown in FIG. Have. Further, as shown in FIG. 2B, a layout information extracting unit 22 for extracting layout information of an area based on a layout database 23 created in advance separately from the area identifying unit 21 is used in combination. You may. As shown in FIG. 2C, a character recognition unit 2 that recognizes a character in a specific area based on a character recognition dictionary 25 created in advance and extracts character information as character information.
4 may be used in combination.

Further, as shown in FIG. 2D, a bar code recognizing unit 2 for recognizing a bar code in a specific area based on the bar code dictionary 27 and extracting it as bar code information.
6 may be used in combination. In addition,
2A to 2D, the area identifying unit 21, the layout information extracting unit 22, the layout database 23, the character recognizing unit 24, the character recognizing dictionary 25, and the barcode recognizing unit 26.
The barcode dictionary 27 and the barcode dictionary 27 may be appropriately combined to form the area information extraction unit 2, or may be combined and used as the area information extraction unit 2.

Next, the operation of the document management system according to the present embodiment will be described. FIG. 3 is a flowchart illustrating a processing procedure of an attribute adding process of an input document image by the document management system according to the first embodiment. First, when the image input unit 1 inputs a document image (step 10), the area information extracting unit 2 executes an area identification process of the input document image (step 11). Here, the area identification processing by the area information extraction unit 2 will be described. For example, the input document image is compressed into a binary compressed image, a rectangle circumscribing the black pixel connected component in the binary compressed image is extracted, and the rectangle is defined as a character rectangle and other rectangles. And extracting a character area and other areas by integrating the rectangles. It is assumed that the rectangles extracted here include rectangles having various attributes such as a character string of one character or two or more characters, ruled lines, tables, figures, and graphs. In this way, the document image is divided into a text area, a table area, a ruled line, a figure area, and the like. By adding information on the type and number of these areas to the document image as attributes, it becomes possible to search later in the document management system using, for example, the number of figures and tables included in the document image. ing.

Next, the layout information extracting section 22 extracts information on the layout from the document image divided into areas by the area identification processing based on the layout database 23 (step 12). The layout information extraction unit 22 extracts the layout information of the document image by, for example, dividing the area-identified document image into smaller areas and lines, and coordinates the positions of the elements such as the respective areas and lines and the corresponding areas. , The size of characters included in a line, the number of lines, and the like are obtained as characteristics of layout information. What is detected as the coordinate value of each element (area, line, number of characters, etc.) by the layout information extracting unit 22 becomes the layout information of the document image. In this way, by extracting the layout information of each area and adding the attribute of the document image, it is possible to search later in the document management system using the layout of, for example, sentences, figures, and tables. I have.

After the layout information is extracted (step 12), character recognition is performed by the character recognizing unit 24 on an area at a specific position (step 13). At the same time as character recognition, barcode recognition is executed by the barcode recognition unit 26 (step 14). Further, only barcode recognition may be performed without performing character recognition. The information read by the character recognition or the barcode recognition is added to the document image as an attribute (step 1).
5) Save the sentence image (step 16). Character strings and barcodes useful as information related to the document image are often attached to specific areas such as the heading, header, and footer of the document image, and extracting the information and managing the document image added as attributes Thus, a document image can be searched in the document management system by using this attribute later.

As described above, in this embodiment, by automatically extracting area information from a document image and adding it as an attribute, an operator does not manually input a document image that is not a fixed form. It is possible to realize a document management system that allows easy retrieval by attributes. In the document management system according to the present embodiment, since area information that can be automatically extracted from an input document image is added as an attribute value, an operator can add attribute values to each image when registering a document image. Can be greatly reduced. Further, in the document management system of the present embodiment, as the area information automatically extracted, the type of area such as a character area, a table area, a ruled line, and a figure area and the number of each area are added as attributes of the document image. Thus, the search can be easily executed by a simple process.

According to the document management system of the present embodiment, since layout information is used as area information, it is possible to add a document image attribute value that cannot be distinguished only by the type and number of areas, and to obtain more detailed information. Can be used to search for a document image. Further, according to the document management system of the present embodiment, since character information described in a specific area is used as area information, it is possible to add an attribute value of a document image that cannot be distinguished only by layout information, A document image can be searched with more detailed information. Further, according to the document management system of the present embodiment, since barcode information described in a specific area is used as area information, it is possible to add a document image attribute value that cannot be distinguished only by layout information. Thus, a document image can be searched with more detailed information.

FIG. 4 is a diagram showing a schematic configuration of a document management system according to the second embodiment. The document management system according to the second embodiment further includes a document image classification table 5 having area information and a document image type defined in advance in the document management system according to the first embodiment. Hereinafter, the same parts as those in the first embodiment will be described using the same numbers. The document image management unit 3 adds the extracted area information to the document image as an attribute, classifies the document image by type, and stores it in the document image storage unit 4.

FIG. 5 is a flowchart showing a processing procedure of an attribute adding process of an input document image by the document management system according to the second embodiment. First, when the image input unit 1 inputs a document image (step 20), the area information extracting unit 2 executes an area identification process for the input document image (step 21). Next, the layout information extraction unit 2
2 extracts information on a layout from the document image divided into regions by the region identification processing based on the layout database 23 (step 22). After extracting the layout information (step 22), the character recognition unit 24 performs character recognition on an area at a specific position (step 23). At the same time as character recognition, barcode recognition is executed by the barcode recognition unit 26 (step 24). Further, only barcode recognition may be performed without performing character recognition.

After layout information extraction, character recognition, and barcode recognition (steps 22 to 24), comparison with the document image type in the document image classification table 5 is performed (step 25). In the document image classification table 5, it is assumed that a document image is classified in advance by combining one or more pieces of area information, and a table for storing the classified document images is created. FIG. 6 is a diagram illustrating an example of the document image classification table. FIG. 6 shows two document image types as an example. The document image type number “001” represents a document image having area information in which the number of figures is 1, the number of tables is 2, and the number of character areas is 4.
The document image type number “002” is the fourth layout information whose layout information is registered in the layout database 23, in which a character string “report” is stored in the character area 1, and This indicates that the character string "lab" exists.

After comparing the area information extracted from the input document image with the document image classification table 5, the type of the suitable document image is determined (step 25). An attribute is added to the image (step 26). After adding attributes (Step 2
6) Classify the sentence images (step 27) and store them in the document image storage unit 4 (step 28). As described above, according to the document management system of the present embodiment, search and classification can be executed even for a document image that is not a fixed form.

For example, the top page of the published patent publication has a barcode area at the upper right, followed by a character string area from the top, such as the heading, symbol, information on the application, the title of the invention, and the abstract, In the lower right is the region of the representative figure. The area information included in the first page of the published patent publication is the same, but these positions may be shifted depending on the amount of information such as the number of IPC classification symbols and the number of inventors.
The system does not have a fixed form, and the system may not be able to easily classify a document by classifying a document image in the fixed form. In addition, even if the document image is not a fixed form as disclosed in the patent publication, the region information included in the case of document images of the same type is often the same. In such a case, the document images can be classified as the same type by defining the document image classification table 5 using the area information whose information amount does not change as in the present embodiment.

As described above, in this embodiment, document images that are not fixed forms but have the same contents can be automatically classified as being of the same type. The search target can be limited for each document image type, and an efficient document management system can be realized. In the document management system according to the present embodiment,
Since the document images are classified according to the combination of the area information, the management, classification, search, and the like of the documents of the same type can be easily executed.

[0022]

According to the first aspect of the present invention, the attribute adding means for adding the area information extracted by the area information extracting means to the document image as an attribute is the same as the document image to which the attribute is added by the attribute adding means. Since document image storage means for classifying and storing attributes is provided for each attribute, it is possible to greatly reduce an operator's work of registering a document image such as adding an attribute value to each image.

According to the second aspect of the present invention, the area information extracting means also extracts the type and number of each area such as a document area, a table area, and a drawing area which are extracted as area information. Since the type and number of areas are also added as attributes,
It is possible to add an attribute value of a document image which is easily searched by a simple process.

According to the third aspect of the present invention, the area information extracting means extracts the position of each area extracted as the area information on the paper as layout information, and the attribute adding means adds the layout information as an attribute. Therefore, an attribute value of the document image that cannot be distinguished only by the type and the number of the regions can be added, and the document image can be searched with more detailed information. According to the fourth aspect of the present invention, the area information extracting means extracts characters contained in a predetermined area of the area information as character information, and the attribute adding means adds the character information as an attribute. An attribute value of the document image that cannot be distinguished only by the layout information can be added, and the document image can be searched with more detailed information. According to the fifth aspect of the present invention, when a bar code is added to the document image, the area information extracting means extracts the bar code as bar code information, and the attribute adding means adds the bar code information as an attribute. Therefore, an attribute value of the document image that cannot be distinguished only by the layout information can be added, and the document image can be searched with more detailed information.

According to the present invention, there are provided two area information extracting means for extracting area information such as a document area, a character area, a table area and a drawing area from a document image obtained by the document image obtaining means. Document image classifying means for classifying the types of document images for each combination of the area information, attribute adding means for adding the area information extracted by the area information extracting means to the document image as attributes, and document image classifying means Document image storage means for classifying and storing document images to which attributes have been added by the attribute adding means for each type based on the classification according to, so that documents of the same type can be managed, and A simple document image management system can be realized.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a schematic configuration of a document management system according to a first embodiment.

FIG. 2 is a diagram illustrating an area information extraction unit.

FIG. 3 is a flowchart illustrating a procedure of a process of adding an attribute of an input document image by the document management system according to the first embodiment;

FIG. 4 is a block diagram illustrating a schematic configuration of a document management system according to a second embodiment.

FIG. 5 is a flowchart illustrating a processing procedure of an attribute adding process of an input document image by the document management system according to the second embodiment.

FIG. 6 is a diagram illustrating an example of a document image classification table.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Image input part 2 Area information extraction part 3 Document image management part 4 Document image storage part 21 Area identification part 22 Layout information extraction part 23 Layout database 24 Character recognition part 25 Character recognition dictionary 26 Barcode recognition part 27 Barcode dictionary

Claims (6)

[Claims] 1. A document image obtaining means for obtaining a document image, and area information extracting means for extracting area information such as a document area, a character area, a table area, and a drawing area from the document image obtained by the document image obtaining means. An attribute adding unit that adds the area information extracted by the area information extracting unit as an attribute to the document image; and a document image that classifies and stores the document image to which the attribute is added by the attribute adding unit for each attribute. Storage means;
A document management system comprising: 2. The area information extracting means also extracts the type and number of each area such as a document area, a table area, and a drawing area, which are extracted as the area information. 2. The document management system according to claim 1, wherein the number and the number are added as attributes. 3. The method according to claim 1, wherein the area information extracting means extracts, as layout information, the position of each area extracted as the area information on a sheet, and the attribute adding means adds the layout information as an attribute. 3. The document management system according to claim 1, wherein: 4. The area information extracting means extracts a character contained in a predetermined area from the area information as character information, and the attribute adding means adds the character information as an attribute. Claims 1 and 2,
The document management system according to claim 3. 5. The area information extracting means extracts the bar code as bar code information when a bar code is added to the document image, and the attribute adding means adds the bar code information as an attribute. The document management system according to any one of claims 1, 2, 3, and 4, wherein: 6. A document image acquiring means for acquiring a document image, and area information extracting means for extracting area information such as a document area, a character area, a table area, and a drawing area from the document image acquired by the document image acquiring means. A document image classifying unit that combines two or more of the region information and classifies the type of the document image for each combination of the region information; and adds the region information extracted by the region information extracting unit to the document image as an attribute. And a document image storage unit for classifying and storing, by type, document images to which attributes have been added by the attribute addition unit based on the classification by the document image classification unit. And a document management system.