JP3483919B2 - Slip document information system - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a slip document information system for reading, sorting, storing, retrieving a plurality of kinds of slips (for example, medical fee bills) having different formats and formats.

[0002]

2. Description of the Related Art The format of a document defines the description items and the order of the items, but does not specify the absolute position coordinates and dimensions. The document format not only defines the description items and the order of the items, but also the absolute position coordinates and dimensions thereof. In a conventional optical character recognition device, a document format is preset as format data, and a paper document in which absolute position coordinates and dimensions of each item of the document are defined can be read. However, it was not possible to read a document in which only the form was specified, such as a legal document.

A conventional document information system for reading slips is disclosed in Japanese Patent Application Laid-Open No. 56-11573 (Japanese Patent Application No. 54-87).
No. 338, filed on July 10, 1979, inventor: Masayuki Kagawa, applicant: Fujitsu Limited), and a mark entry area is provided on the slip to indicate items to be entered in advance. There has been proposed a method of reading a predetermined area on a slip by reading a mark written in the area. However, in the conventional method, since the mark writing area is provided, the area of the writing area on the paper surface increases, and there is a problem that the number of input items cannot be increased. In addition, since it is necessary to newly prepare a mark writing area, there is a problem in that various existing slips cannot be read as they are. Furthermore, in the case of a voucher document whose format is determined by law, etc., it is not allowed to provide a mark entry area as in the past because the format is legally defined, and the format document without the mark entry area is read. There was a problem that I could not do it.

Also, the conventional document image file system is
JP-A-59-14070 (Japanese Patent Application No. 57-1234)
No. 60, filed on July 15, 1982, inventor: Koichi Ejiri, applicant: Ricoh Co., Ltd., as described in a halftone image photograph or a binary graphic image bar graph. An image file method has been proposed in which a characteristic parameter is extracted from an area of a document including a text and stored in association with an image of the area. However, in the conventional method, only the identification of the photograph and the bar graph is described, and the document such as the ruled line and the frame composed of the ruled line and the character string is not described. Further, neither the character recognition of the character string of the slip nor the file method of masking the document area according to the degree of confidentiality is considered.

Further, there is a filing system which automatically identifies the type of a document with respect to a fixed input document and gives the same classification code as that of a document already stored, as disclosed in Japanese Patent Laid-Open No. 61-7.
5477 (Japanese Patent Application No. 59-196616, Showa 5)
Application filed on September 21, 1984, inventor: Yasuaki Nakano,
Hiromichi Fujisawa, Masaaki Fujinawa, applicant: Hitachi, Ltd.). However, this conventional method mainly stores images, and considers setting a reading area for character recognition and verifying the validity of the described content based on the data of the recognition result. Absent. For this reason, it is not possible to read the description content and verify the validity of the content according to the type of document.

[0006]

The above-mentioned respective prior arts do not describe that a serial number or a bar code is added to the input paper document itself, and a desired document image is displayed from an accumulated image file. Although it can be printed, it is difficult to search for the original paper document itself, and the work of searching for the original was a work that requires a lot of human hands. In addition, sorting of paper documents themselves was not considered, and a lot of manpower was required for sorting, organizing, and storing the originals.

Further, in each of the above-mentioned prior arts, confidentiality items are not taken into consideration when displaying and printing the stored document images, and, for example, private contents and business important information are described in the documents. Even in such a case, the image of the search result is displayed as it is on the screen, or the display of the entire surface is prohibited for each image unit, and the partial area of the document cannot be viewed according to the confidentiality level.

Further, in each of the above prior arts, it is premised that all input paper documents are stored, and depending on the contents,
It does not take into consideration selection and accumulation. Therefore, a large amount of documents are accumulated, and there is a problem that the capacity of the storage device is exceeded in a small file system.

A first object of the present invention is to identify the types of formats of a plurality of slip documents, automatically set the reading area, perform character recognition, and verify the validity of the described contents.

A second object of the present invention is to sort a paper document by printing a reference number or a bar code on the paper document according to the recognition result.

A third object of the present invention is to extract an area in which confidential contents are described from a document image, mask the image of the confidential area, and display and print the search result.

A fourth object of the present invention is to automatically determine the necessity of storage according to the contents of a document and save the storage area.

A fifth object of the present invention is to provide a document information system which recognizes, sorts and stores data via a network.

[0014]

In order to achieve the first object, the type of style is identified from the document image, the description content is read, and the read result satisfies the condition registered for each entry item. Whether or not it is determined.

In order to achieve the second object, the sorting mechanism and the printing mechanism are controlled according to the recognition result.

Further, in order to achieve the third object, the partial area in which the item is described is automatically extracted from the document image, and a password for access is given to the confidential area. It is a thing.

In order to achieve the fourth object, the description content, for example, the score is read, and the surface image is stored only for the document having the score larger than a certain value.

In order to achieve the fifth object, the image is compressed and transmitted to the network, the recognition result is transmitted, and the sorting station and the storage station connected to the network are controlled by the data. Is.

[0019]

Operation: The surface images of a plurality of types of slips are input to identify the type of slip document format. Then, the reading area is extracted according to the type of style and character recognition is performed. Based on the recognition result data, the validity of the described contents is verified according to the preset judgment conditions. Further, a serial number is printed on the paper document or a bar code is printed according to the recognition result, and the paper document is sorted according to the document type.
Further, an area in which the confidentiality content is described is extracted according to the type of document, and the image of the confidentiality area is masked so that it can be displayed.

[0020]

The present invention will be described in detail below with reference to the drawings. FIG. 1 is a diagram for explaining the configuration of a slip information system which is an embodiment of the present invention.

An input / sorting station 103, a storage station 101, a recognition station 105, a correction station 106, and a search station 107 are connected to the network 100. The input / sorting station 103 is composed of a computer. The station 103 is connected to an input / sorting device 104 for inputting images of sorting / sorting slip documents and sorting of slip documents. The device has a hopper for setting documents, a stacker for storing documents after sorting, and a control unit thereof.
Further, an optical system such as an optical image pickup device for inputting an image and a lighting device is provided. Here, the input image is data-compressed and the image data is sent to the network 100 via the station 103. Further, the recognition result is received from the network 100, and the document is classified by using the recognition result as a classification control signal. The storage station 101 is composed of a computer, and is connected to a mass storage device 102 such as an optical disk or an array disk, and stores the recognition result data and the input image. The storage station 101 can select documents to be stored based on the recognition result. Further, access information is added to the partial area in which the confidentiality content is described, and the partial area is stored in the storage device 102. The recognition station 105 performs sheet identification processing for determining the type of the slip document for the image sent via the network 100, and recognizes the character string described in the item. The correction station 106 interactively corrects the recognition result using a screen. Here, the reading area is enlarged and displayed on the screen to facilitate the correction. The search station 107 can search for the slip image stored in the storage station 101 via the network, and can search for a desired slip based on the content described in the item and display it on the screen. At this time, access information is given to the partial area in which the confidentiality content is described, and the confidentiality item can be masked and displayed depending on the access level of the user.

FIG. 2 is a diagram for explaining the processing flow of the slip document. As an example of the slip document, a processing flow of a medical fee bill (receipt) is shown. Enter the data for the main items described in the receipt and confirm the qualification record with a computer. In this processing flow, first, in step 200, the receipt is set in the hopper for input. Next, in step 201, a receipt sorting process is performed. Here, as the types of receipts, the outpatient receipt 210 and the inpatient receipt 211 are classified according to their styles. In step 202, initial information is input. Here, as the initial information, the distinction between the own prefecture and other prefectures is input as the regional attribute for performing the receipt process for each prefecture. In addition, the initial values of the office name, medical treatment category, and reference number for processing the receipt are registered. For the hospitalized receipt 211, the reference number is assigned in step 208. Next, reading is executed in steps 203 and 204. Step 203 is
This is a batch reading process that performs batch processing of slips, and stores normal and unread (rejected) recognition results in another stacker and prints a reference number. Step 204 is a process of sequential correction reading, in which the recognition result is interactively corrected via the correction screen for the reject rejected as a result of the recognition. The normal data recognized in step 203 and the normal data corrected in step 204 are stored in a floppy disk or the like. Further, the rejected reject in step 203 is stored in another stacker. In addition,
A punching operation is performed in step 209 by collecting the hospitalized portion receipt and the rejected portion in step 203. Machine processing such as qualification record confirmation processing is performed as shown in step 207 based on the medical care contents described in the receipt. here,
Only the receipts in which expensive points are described are selected and stored in the optical disc in step 205. Step two
In the qualification record confirmation processing of 07, the receipt 206 stored in the optical disk is searched 206 in step 205, and a desired receipt is displayed on the screen to confirm the qualification record.

FIG. 3 shows image input, storage, retrieval of a slip document,
It is a processing flow of the system which inspects the contents. In step 300, the front side image of the slip document is input, and in step 30
In step 1, sheet identification processing is performed to identify the type of slip format. The detailed processing flow of the sheet identification will be described later with reference to FIG. Next, in step 302, a reading area is extracted and set according to the type of the style. The detailed processing flow of the reading area setting step 302 will be described later with reference to FIG. In step 303, character recognition is performed in the reading area, and in step 304, the validity of the described contents is verified based on the data of the recognition result. The detailed processing flow of step 303 will be described later with reference to FIG.

FIG. 4 is a diagram showing a detailed processing flow of the sheet identification processing in step 301. Here, a frame and a character line are extracted from the input image, and matching is performed with the sheet identification dictionary 430 registered in advance from a standard document. First, in step 400, minute noise is removed from the input image. Next, in step 401, a ruled line is extracted using a run that is a line segment in which black pixels are continuous, and step 40
In step 2, the tilt angle of the ruled line is calculated. In step 403, the image is rotated by the tilt angle to correct the rotation. Then, again, in step 404, the ruled line is extracted. FIG. 6 is an explanatory diagram of a process of extracting a horizontal ruled line and a vertical ruled line from an image by this series of processes and calculating start point coordinates and end point coordinates thereof. 601 is a result of extracting horizontal ruled lines 602, 603, and 604 from the slip image 600. Further, 610 is the result of extracting the vertical ruled lines 611 to 616. In step 405, a character line, which is a connection of characters, is extracted in the shape of a rectangle, and the coordinates of its corners are obtained. In order to correct the horizontal and vertical misalignment of the document, a reference point of the document is determined in step 406. In step 407, the coordinates of the intersections corresponding to the corners of the frame are calculated from the horizontal ruled lines and the vertical ruled lines. Since each ruled line can be expressed by a straight line type, it is possible to easily obtain the intersection coordinates of the straight line. In step 408, the obtained intersection coordinates are normalized in size. Further, in step 409, a frame is generated based on the intersection. FIG. 7 is a diagram for explaining a process of obtaining an intersection from horizontal and vertical ruled lines and then generating a frame.
Horizontal ruled lines 602-604 and vertical ruled lines 611-61
Based on 6, the intersection points 710 to 723 are calculated. Then, the intersections are sequentially traced to generate frames 701, 702 to 706. In step 410 of FIG. 4, frame matching is performed. At this time, matching with a plurality of standard frames stored in the sheet identification dictionary 430 is performed. The dictionary 4
The position of the frame and the position of the character line are registered in advance in 30, and the registration flow is as follows. First, in step 420, the standard slip image is input, and then in step 421, the frame and character line to be input are displayed. Specify the position from the screen. In step 422, the position and rotation are corrected and registered as a dictionary. Further, in step 411, character line matching is performed. At this time, matching with a plurality of standard character lines stored in the sheet identification dictionary 430 is performed. Details of character line matching will be described later with reference to FIG.
The frame matching 410 and the character line matching 41 described above
In accordance with the result obtained in step 1, the sheet type number of the most similar slip among the standard slips is determined in step 412. Further, in step 413, the generated frame position and character line position are output. Further, in step 414, the sheet type number is output.

FIG. 5 shows a recognition coordinate system having displacement, rotation and expansion / contraction. The slip sheet 500 has a frame ruled line 506. The recognition coordinate system is represented by a recognition coordinate system (Xr, Yr) 511 without expansion and contraction and a recognition coordinate system (Xm, Ym) 512 with expansion and contraction. These coordinate systems are
It is an observation coordinate system consisting of a sensor and an optical system. Further, the format data generating coordinate system (Xf, Yf) 513 is a coordinate system based on the edge of the sheet, and is a coordinate system representing the dimensions of the actual product. The target point (xr, yr) 507 in the sheet expressed in the recognition coordinate system without expansion and contraction is the origin 510 of the recognition coordinate system and the origin 509 of the format data generation coordinate system.
It can be expressed by the shift amount (dx, dy) and the rotation angle Θ.

[0026]

## EQU1 ## xr = dx + xf.cos.THETA. + Yf.sin.THETA. Yr = dy-xf.sin.THETA. + Yf.cos.THETA. Where (xf, yf) are the coordinates of the target point 507 in the format data generation coordinate system. Furthermore, the target point 507 in the sheet expressed by the elastic recognition coordinate system
(Xm, ym) is the expansion / contraction rate mx, my,

[0027]

[Formula 2] xm = mx × xr ym = my · yr Then,

[0028]

## EQU00003 ## xm = m (dx + xf.cos.THETA. + Yf.sin.THETA.) Ym = m (dy-xf.sin.THETA. + Yf.cos.THETA.) When reading in a sheet, the format data is converted according to the above equation, the image is accessed, and character recognition is performed. In addition, the reference point for alignment 508
The value of the recognition coordinate system of is the recognition coordinate system (xr0, y
r0) and the recognition coordinate system (xm0, ym0) with expansion and contraction. In addition, the value of the coordinate system for format data generation is (x
f0, yf0). The positional deviation amount (dx, dy) is

[0029]

## EQU00004 ## dx = (1 / mx) .xm0-xf0.cos
Θ-yf0 · sin Θ dy = (1 / my) · ym0 + xf0 · sin Θ-yf
It can be represented by 0 · cos Θ. The rotation angle Θ is detected based on the inclination of the ruled line. Next, the expansion / contraction ratios mx and my are calculated.
For example, the width and height of the outermost black line frame are detected and calculated by comparing with the corresponding width and height of the format data. Further, the position (xm0, ym0) of the alignment reference point in the recognition coordinate system is obtained and compared with the position (xf0, yf0) of the point preset in the format data (expressed in the format data generation coordinate system). Then, the positional deviation amount (dx, dy) is obtained. In sheet identification, the coordinates of the target point (x
m, ym) is the coordinate system (xf,
yf), and compares this with the coordinate data in the format of the sheet identification dictionary.

FIG. 29 shows a method of expressing the relative positional relationship of character lines. Now, with respect to the character line 2900 of interest, a positional relationship in eight directions is expressed. If the directions are expressed as north, south, east, and west, there is a character line 2911 in the north direction 2901 of the character line 2900 of interest. In addition, the character line 2915 is in the south direction 2905, and the character line 2913 is in the east direction 2903.
However, there are character lines 2917 in the west direction 2907. Furthermore, the northeast direction 290 of the character line 2900 to be noticed
2, a character line 2912, a northwest direction 2908, a character line 2918, a southeast direction 2904, a character line 2914,
There are character lines 2916 in the southwest direction 2906, respectively. In this way, a character line having a positional relationship in eight directions with respect to the focused character line is expressed.

FIG. 8 is a diagram for explaining the character line matching processing method in step 411. Examples of partial areas of the character lines of the slip document are shown in 810, 811, and 812. In the character line area 810, the extracted character lines 801, 802, 80
As shown in Fig. 3, the relative positional relationship between them is a vertical north-south relationship. Now, assuming that the character line of interest is 801, the character line 802 is 8 in the south direction with respect to the character line 801.
20. In addition, for the character line 802, the character line 80
3 is also in the south direction 821. On the other hand, in the example of the character line area 811, the relative positional relationship between the character lines 804, 805, and 806 is a left-right east-west relationship. The character line 805 is in the east direction 822 with respect to the character line 804, and the character line 8
For 05, 806 is also eastward 823. Further, in the example of the character line area 812, there is a diagonal positional relationship, and the character line 808 is in the northeast direction 824 with respect to the character line 807. In addition, the character line 80
9 is also in the northeast direction 825. Step 411
In the character line matching, the relative positional relationship of such character lines is registered in the sheet identification dictionary 430 in advance, and it is determined whether or not the relative positional relationship of the character lines extracted from the input image matches. judge.

FIG. 9 is a detailed processing flow of the reading area setting processing shown in step 302 of FIG. In step 900, the sheet type number obtained in step 301 of sheet identification is input. Next, in step 901, the process proceeds to each reading area, and in step 902, the corresponding reading format parameter is read from the format parameters 910 registered for each sheet type. Here, the format parameters include the character type, the number of characters, and the position of the reading field. In step 903, the reading area is calculated based on the reading field position read from the format parameter and the position shift amount from the reference point position obtained in step 406.

FIG. 10 is a detailed processing flow of the character recognition processing in the reading area shown in step 303. For each read field in step 1000, step 100
The processing from 1 to 1004 is performed. Step 1001
Then, the character line in the reading area is extracted. Step 10
In 01, unnecessary characters printed in the reading area are removed in advance. Then, the character line extracted in step 1003 is separated into character units. In step 1004, character recognition is performed for each character. This character recognition is
It can be easily realized using conventional techniques.

FIG. 11 is a diagram showing a processing flow of the slip document information system for reading a slip document and selecting and storing the slip document based on the read result. Step 11
The processing from step 1101 onward is repeated until a document is attached at 00. In step 1101, an image of the slip document is input. Then, in step 1102, step 301
A sheet identification process similar to that is performed. Then step 1
Similarly, in step 103, the reading area is extracted, and step 11
At 04, character recognition is performed in the reading area. In step 1105, it is determined whether the numerical data of the recognition result is within a preset range. If it is within the preset range, the image is stored in a storage device such as an optical disk in step 1106. If it is not within the set range, the image is not stored in the storage device. The above-described series of processing is performed in step 1100 until the document is exhausted. It is not necessary to store all the paper documents input by this processing, and it is possible to select and store depending on the description content. Therefore, there is an effect that the storage capacity can be saved in a small-scale file system.

In a document information system for inputting, storing, retrieving, and checking the contents of a slip document, the confidential area image is masked and displayed according to the access level of the user. FIG. 12 shows a processing flow for registering the confidential area from the slip document image. An image is input in step 1200, and sheet identification processing 301 is performed in step 1201. Then, in step 1202, the confidential area is extracted. Here, the sheet type and the position of the area extracted in step 1201 of the previous sheet identification are output, and it is determined for each area whether or not the area is a confidential area. At this time, the sheet identification dictionary 430 registered in advance is read to determine whether the area is a confidential area. The dictionary 430 has a confidentiality level for each area indicated by a frame. In step 1203, the coordinates of the confidential area are stored in the storage device together with the image data.

FIG. 13 is a flow chart of a process of masking and displaying the image of the confidential area according to the access level of the user with respect to the slip image. First, in step 1300, the password of the user is input. Then, in step 1301, the access level of the user is set using the user access level table 1600 shown in FIG. FIG. 16 is an explanatory diagram of the user access table. User name 160
1, access level 16 for password 1602
03 is stored, and the access level corresponding to the user is set. Next, the processing of steps 1303 and 1304 is repeated until the area is formed in step 1302. In step 1303, the access level set for each area is read from the area access table 1500. FIG. 15 shows the area access table. The table 1500 has an access level 15 for the standard slip document for each identification number 1501 assigned to the area.
02 is assigned. Then, in step 1304, it is determined whether the access level of the user is stronger than the access level of the area. If the access level of the user is strong, the masking of the area is released in step 1305. On the contrary, if the access level of the area is stronger than the access level of the user, a mask is set in the area in step 1306. The above series of processing is repeated until a region is formed in step 1302. Then, in step 1307, the slip image is displayed. Here, the display process is executed for each area, and the partial image in which the mask is set in the area is not displayed as it is, but for example, an image filled with a black image is displayed. The partial image in which the mask of the area is released is displayed as it is. In this embodiment, when displaying and printing the stored document images, access levels are given to confidential items. For example, when the document contains private contents or business-important information, There is an effect that the partial area of can be seen according to the confidentiality level and the confidentiality can be maintained.

FIG. 14 shows an example of displaying a slip document image. Areas 1401, 1402, 1403, 1404, 14 corresponding to each item of the slip on the display screen 1400
15, 1405, 1406, 1407, 1408, 14
09 is set. At this time, the shaded portions 1410, 1411, 1412, 14 are displayed according to the access level of the user.
In 13 and 1414, a mask is set for confidentiality, and the area is filled with black.

FIG. 17 shows a processing flow for printing a barcode on a slip document according to the result of reading the document. In step 1700, a document image is input. In step 1701, sheet identification for identifying the type of slip document is performed on the input image, and in step 1702, an area in the sheet is read by character recognition. Then, in step 1703, the barcode is printed on the front surface of the paper document according to the recognition result.

FIG. 18 shows a processing flow for searching a document by a bar code. When the reference number of the document to be searched is input in step 1800, a bar code corresponding to the reference number is set in step 1801. Then, in step 1802, steps 1803 and 18 are performed until the document is exhausted.
The processing of 04 and 1805 is repeated. First, step 18
In 03, paper feeding of the slip document is executed, and the bar code printed on the paper surface is read in step 1804. Next, in step 1805, it is determined whether the read barcode matches the barcode set in step 1801. If they match, it is determined that the document is a desired slip document and is stored in the selected stacker in step 1806. If they do not match in step 1805,
In step 1807, the sheets are stored in a normal stacker, and in step 1802, the process is repeated until a document is attached. Since the barcode is printed on the original according to the present embodiment, not only the desired document image is displayed and printed from the accumulated image files, but also the original paper document itself can be easily searched using the barcode. There is an effect that can be. Figure 1
9 is a processing flow for identifying the types of slip documents and sorting. The slip document image is input in step 1900, and the sheet is identified in step 1901 by the sheet identification dictionary 1910. Then, in step 1902, the in-sheet reading is executed. At this time, the previously registered format parameter 1911 and the character recognition dictionary 1912
To use. In step 1903, the slip documents are sorted according to the sheet type that is the result of sheet identification. According to the present embodiment, it becomes possible to automatically sort the slip documents, which was conventionally done manually, and it is possible to reduce the amount of work done manually.

FIG. 20 shows input / output connected to the network.
It is a figure explaining operation control of a sorting station and a recognition station. An input / sorting station 2002 and a recognition station 2001 are connected to the network 2000. To control the process, first, the recognition station 2001 sends an activation of the entire system to the input / sorting station 2002. Input / sort station 2
002, the slip document set in the hopper upon receiving this activation command inputs an image with the scanner of the input device 2004.
The image is data-compressed in the image processor in the input device 2004 and transferred to the recognition station 2001 via the network 2000. The transferred image is decompressed and the sheet is identified and read. The recognition result is transferred to the input / sorting station 2002 via the network 2000, and the stacker of the sorting apparatus 2005 is selected. The above process is repeated until there is no slip document for which the hopper is set and the hopper becomes empty. When the stacker becomes full on the way, a warning sound is generated and the stacker is temporarily stopped. When the hopper becomes empty, the input / sorting station 2002 transmits an end command to the recognition station 2001, and the processing of the entire system ends.

FIG. 21 is a flow chart of a slip document process in which the types of slip documents are identified, characters are recognized, and a serial number is assigned to a paper document according to the recognition result to perform sorting. Step two
At 100, the image of the slip document is input. Step 2101
The sheet identification for identifying the slip type is executed with. In step 2102, the inside of the sheet is read, and the reference number is assigned in step 2103 according to the recognition result. In step 2104, the documents are sorted according to the type of the slip or the description content and stored in a predetermined stacker.

FIG. 22 shows an example of a slip document with a reference number. In the slip document 2200, the reference number 2201 is numbered. Since the position of the edge of the document is detected, the reference number can be numbered at the preset coordinate position. Also, the ink color for the printing number can be switched to the designated color.

FIG. 23 is a diagram showing the structure of input / sorting. A slip document 2320 is set in the hopper 2300. A computer 2304 controls the entire input / sorting process. It controls communication with the network 2321. The hopper control unit 2310 controls the hopper mechanism unit 2315 and sequentially feeds the paper to the sensor unit 2313. The sensor control unit 2311 digitizes the image signal, and the computer 230
Send to 4. 2314 is a hitting number mechanism, and the hitting number control unit 23
The reference number sent from the computer 2304 by 12 is printed. Sorting mechanism unit 2316, 2317, 231
The computer 8 is controlled by the command of the computer 2304 to store the sent slip document in a predetermined stacker.
The selected slip documents are sequentially stored in the stackers 2301, 2302 and 2303. Each stacker is equipped with sensors 2307, 2308, and 2309 that detect the fullness of the document, and the fullness detection sensor control unit 2306 sends a stacker fullness signal to the computer 2304. 23
Reference numeral 05 is a warning sound generation unit, which generates a warning sound when the stacker is full and the computer 2304 suspends the processing.

FIG. 24 is an example of a screen transition diagram of the slip document information system. Main screen 2 at system startup
401 is shown in FIG. The main screen 2500 includes a selection item 2501 and a selection item input prompt 2502. 2 by the input method designation screen 2402
It is possible to switch the type of input, one of which is the scanner input screen 2403 and the other is the file input screen 24.
Perform at 04. An image display screen 2405 allows an image to be input by a specified input method, and the input image can be displayed and confirmed on the screen. From this screen, it is possible to return to the main screen 2401 and transit to, for example, the style / form identification screen 2406. Here, as shown in FIG. 26, the extracted frame 2601 can be displayed on the screen 2600. Further, as shown in the screen 2602, the extracted frame 2603 and the frame 2604 provided as a dictionary can be displayed in an overlapping manner. Reference numeral 2408 denotes a recognition result screen, on which a document image and a recognition result can be displayed at the same time and correction can be performed. FIG. 27 is a diagram illustrating the recognition result screen. An upper image display area 2701 and a lower image display area 2702 are enlarged and displayed on the slip document 2700. On the recognition result screen 2710, the upper image is enlarged and displayed at 2711. Also, the lower image is enlarged and displayed in 2712. The recognition result is displayed in the portion 2713, and the correction can be performed interactively. FIG. 28 is a diagram illustrating another example of the recognition result screen. In this embodiment, the recognition result screen has two screens,
An upper image 2801 is displayed on the first screen 2800, and the recognition result is displayed in the portion 2802. Further, a lower image 2804 is displayed on the second screen 2803, the recognition result is also displayed on 2805, and the correction is performed interactively.

FIG. 30 is a detailed processing flow of the adequacy test in step 304, which is an example for the receipt.
At step 3000, the gender and the content of the request are verified.
Some medical procedures differ depending on the sex, and it is verified whether the contents of the request for inspection, surgery, etc. correspond to the sex. Further, in step 3001, the age and the content of the bill are verified. Here, there are restrictions on the content of dispensing and surgery depending on the age, and it is verified whether the content of the claim is appropriate for the age. Further, in step 3002, the age and slip format are verified. Here, by age,
For example, the form of the slip is prescribed in advance by a newborn baby, an infant, an old man, etc., and it is verified whether or not the form of the slip obtained by sheet identification matches the stated age.
In step 3003, it is determined whether the insurance card number matches the registered name. Further, in steps 3004 and 3005, the actual medical treatment days and the total points are verified.

[0046]

As described above, the type of style is identified from the document image, the description content is read, and it is determined whether or not the read result satisfies the condition registered for each entry item. Can be read. Further, the reading area is automatically set, the characters are recognized, the validity of the described contents is verified, and the described contents of the slip can be automatically inspected. According to the recognition result, the sorting mechanism and the printing mechanism are controlled. According to the recognition result, the reference number or the bar code can be printed on the paper document, and the paper document can be sorted automatically, which reduces the sorting work. effective. Since a password for access is given to the confidential area, the image of the confidential area can be masked to display and print the search result, which has the effect of improving the reliability of the system. The description content, for example, the score is read, and the surface image is stored only for the document having the score larger than a certain value, which has an effect of saving the storage area.

[Brief description of drawings]

FIG. 1 is a diagram illustrating a configuration of a slip information system.

FIG. 2 is a diagram illustrating a processing flow of a slip document.

FIG. 3 is a diagram showing a processing flow of a system for performing image input, storage, search, and content inspection of a slip document.

FIG. 4 is a diagram showing a detailed processing flow of sheet identification processing.

FIG. 5 is a diagram for explaining a recognition coordinate system having displacement, rotation, and expansion / contraction.

FIG. 6 is an explanatory diagram of a process of calculating ruled line coordinates.

FIG. 7 is a diagram illustrating a process of calculating a ruled line intersection and generating a frame.

FIG. 8 is a diagram illustrating a processing method of character line matching.

FIG. 9 is an explanatory diagram of a detailed processing flow of reading area setting processing.

FIG. 10 is an explanatory diagram of a detailed processing flow of character recognition processing in a reading area.

FIG. 11 is an explanatory diagram of a slip document information system that selectively stores a slip document.

FIG. 12 is an explanatory diagram of a processing flow for registering a confidential area from a slip document image.

FIG. 13 is an explanatory diagram of a processing flow in which an image of a confidential area is masked and displayed according to a user access level.

FIG. 14 is a diagram illustrating an example of displaying a slip document image.

FIG. 15 is an explanatory diagram of an area access table.

FIG. 16 is an explanatory diagram of a user access level table.

FIG. 17 is an explanatory diagram of a processing flow of printing a barcode on a document using a recognition result.

FIG. 18 is an explanatory diagram of a processing flow for searching for a document using a barcode.

FIG. 19 is an explanatory diagram of a processing flow for identifying the types of slip documents and sorting.

FIG. 20 is an explanatory diagram of operation control of a station connected to the network.

FIG. 21 is an explanatory diagram of a slip document processing flow in which a reference number is assigned and sorting is performed according to a recognition result.

FIG. 22 is a diagram showing an example of a slip document with a reference number.

FIG. 23 is a diagram showing a configuration of input / sorting.

FIG. 24 is a screen transition diagram of the slip document information system.

FIG. 25 is a diagram showing a main screen.

FIG. 26 is a diagram showing a frame display screen.

FIG. 27 is a diagram illustrating a recognition result screen.

FIG. 28 is a diagram illustrating another example of the recognition result screen.

FIG. 29 is a diagram illustrating a relative positional relationship between character lines.

FIG. 30 is a detailed processing flow of a validity test.

[Explanation of Codes] 100 ... Network, 101 ... Storage Station, 1
05 ... Recognition station, 103 ... Input / sort station, 301 ... Sheet identification processing step, 304 ... Validity verification step, 410 ... Frame matching step, 4
11 ... Character line matching step, 910 ... Format parameter, 1306 ... Area mask setting, 1703
... Bar code printing step, 1804 ... Bar code reading, 2103 ... Serial number numbering, 2104 ... Sort processing.

─────────────────────────────────────────────────── ─── Continuation of front page (72) Inventor Masafumi Koga 1-280, Higashi Koikeku, Kokubunji, Tokyo Inside Central Research Laboratory, Hitachi, Ltd. (72) Kazuki Nakajima 1-280, Higashi Koikeku, Kokubunji, Tokyo Hitachi, Ltd. Central Research Institute (72) Inventor Shozo Kadota 2880 Kokufu, Odawara, Kanagawa Hitachi Storage Systems Division (72) Inventor Kiyomichi Kurino 2880, Kokufu, Odawara Kanagawa Storage Systems Division (72) Inventor Kenyuki Sugimoto 2880, Kokufu, Odawara, Kanagawa, Ltd. Storage Systems Division, Hitachi, Ltd. (56) References JP-A-4-268685 (JP, A) JP-A-2-56086 (JP, A) JP-A-5-244150 (JP, A) JP-A-3 148785 (JP, A) JP 58-78263 (JP, A) JP 64-26993 (JP, A) JP 61-75477 (JP, A) JP 5-258099 (JP, A) (58) Fields surveyed (Int.Cl. ⁷ , DB name) G06K 9/00-9/82 G06F 19/00

Claims (6)

(57) [Claims] 1. A enter the voucher document as an image, the accumulation of the image search, the contents inspected, and the slip document information system for selection of the document document, means for storing the type of style in advance as a dictionary data A means for extracting a frame and character lines from the input slip document, a means for identifying the format type of the slip document by collating the dictionary data with the extracted frame and character line data, A means for extracting a reading area according to the type, a means for recognizing a character by setting the reading area, a means for testing the validity of the described contents based on the data of the character recognition result, and a means for checking the validity test result. And a means for selecting the slip documents based on the slip document information system. 2. A storage unit for storing an access level for each area in the slip document, a unit for reading out an access level corresponding to an input user ID, and the access level read out for each area. 2. A unit for comparing with an access level, a unit for masking the image of the confidential area and displaying it on a display unit, and a unit for releasing and setting the mask according to the comparison result. Described slip document information system. 3. A means for setting a plurality of areas of a slip document as a display area, a means for simultaneously displaying an image of the display area and a recognition result of the recognition means on a screen, and modifying the recognition result via the screen. 3. A reading area extraction unit that enlarges and displays an image of the reading area, wherein the reading area extracting unit extracts the reading area from the set plurality of areas. Described slip document information system. 4. The validity checking means judges whether or not the numerical data of the recognition result of the character recognizing means is within a preset range, and the slip document information system is further within the range. 4. The voucher document information system according to claim 1, wherein a voucher document image determined to be selected is stored, and only the selected image is stored. 5. A voucher document information system according to any one of claims 1 to 4, further comprising means for sorting the documents documents based on the type of the identified documents document. 6. A document information system for sorting slip documents via a network, an input station for inputting a front image of a slip document, a recognition station for identifying a type of the input image, and a recognition result of the recognition station. According to the above, a sorting station for sorting the paper document and a storage station for storing the image are provided, the recognition station stores the type of style as dictionary data in advance, and extracts a frame and a character line from the input specification document. Means, means for identifying the style type of the slip document by collating the dictionary data with the extracted frame and character line data, means for extracting a reading area according to the style type, and a reading area And the means for character recognition and the validity of the described contents is verified based on the character recognition result data. And stage, based on the validity test result
A means for selecting the slip document, image data via the network, data on the type,
And a document information system characterized by transmitting sorting control information.