JP2007133593A - Discrimination system and discrimination program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To discriminate the sort of a document entered by an electronic pen. <P>SOLUTION: A server 104 divides documents 100 prepared by predetermined sheets in a predetermined area and stores an area definition master 126 in which points to be applied when a locus exists are defined in each area. Points of locus information acquired from the electronic pen 101 are totalized in each document by using the area definition master 126. An adaptive rate of the totalized points of the calculated locus information to all points defined in the area definition master 126 is calculated in each document to discriminate the sort of the document in which the locus information is entered. Then the information of the discriminated document sort is displayed on a screen to support judge's document sort judgment. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a technique for identifying a form to be filled in using an electronic pen.

In recent years, electronic pens (digital pens) have been used as input devices, and representative examples thereof include “Anoto (registered trademark) pen” manufactured by Anoto (registered trademark) Corporation.
In this electronic pen, a user writes characters on the paper on which a predetermined dot pattern has been printed, and the camera built in the electronic pen photographs the dot pattern. The pen movement trajectory is detected, and the entry data entered by the user is acquired as the trajectory information as electronic data. The acquired trajectory information can be transmitted to a server or the like through a network or the like.

  For the trajectory information, only the entry data is acquired by the electronic pen, and the characters, layout, etc. of the form printed on the paper itself are not acquired. Therefore, in order to identify what type of form (form type) the entry data is written in, it is necessary to use a dot pattern having a different pattern for each form.

For the identification system for identifying the type of form in which the entry data is written, there are the following techniques.
For example, in Patent Document 1, a type recognition area is provided in a form, and a user enters a check or the like in the type recognition area, thereby supplying trajectory information including data entered in the type recognition area to the server. The server recognizes the type of the form entered by the user based on the trajectory information. The dot pattern printed in the type recognition area is different for each type of form, and the server identifies the form type using this.

Further, in Patent Document 2, a server obtains identification information corresponding to a belt-shaped pattern by providing a belt-shaped pattern (barcode or the like) to which identification information is assigned to a form, and the user traces the position with an electronic pen. The server uses the identification information to identify the form.
JP 2004-127198 A (paragraphs “0007” to “0023”, FIG. 8) JP 2003-345503 (paragraphs “0009” to “0015”, FIG. 9)

However, it may be necessary to provide a plurality of forms regardless of the dot pattern. In some cases, it is necessary to distinguish forms even on paper with the same dot pattern.
The form in Patent Document 1 is provided with a type recognition area, and does not automatically determine the type of form.
In Patent Document 2, specific identification information is assigned to each band pattern in advance, and the user can input the corresponding identification information by tracing the band pattern with an electronic pen. The form type cannot be automatically determined. In other words, it is not possible to identify the form type without imposing a burden on the user, and an operation for identifying the form type is a burden on the user.

  An object of the present invention is to solve the above-described problems and provide an identification system and an identification program for identifying a form type filled in using an electronic pen without burdening the user.

In order to solve the above problems, the present invention analyzes a form created using a predetermined sheet, an electronic pen that generates locus information by writing to the form, and locus information acquired from the electronic pen. In the identification system for identifying the type of the form using a processing device that converts to electronic data, the processing device is an entry defined because the entry position by the electronic pen is different for each type of the form Using the storage means for storing the position definition information, and the entry position definition information and the locus information input from the electronic pen by writing to the form to be identified, the entry position definition information and the locus information By calculating the height of the degree of coincidence, identifying means for identifying the type with the calculated degree of coincidence as a candidate, and displaying the information of the identified type and the trajectory information on the screen And the configuration and output means to a type of the document to judge selectable.
Other means will be described in an embodiment described later.

  According to the present invention, it is possible to identify a form type entered using an electronic pen without placing a burden on the user.

The best mode for carrying out the present invention (hereinafter referred to as “embodiment”) will be described in detail.
In this embodiment, the identification system is abbreviated as “system” as appropriate. The “form type” is a concept including the form type, the layout of the entry area in the form, vertical writing, horizontal writing, etc., and “printing” is a concept including normal printing and printing out by a printer. And

(System configuration)
FIG. 1 is a configuration diagram of an embodiment to which an identification system according to the present invention is applied.
The form 100 is created, the trajectory information entered in the form 100 using the electronic pen 101 is acquired, the server 104 identifies the type of the entered form using the trajectory information, and the result is displayed. The process up to now will be described with reference to FIG. In addition, about each process (S201-S207) shown with a double surrounding line in FIG. 1, it mentions later using FIG.

First, in a printing company, for example, a predetermined format (ruled lines, frames, item names, etc.) is overprinted using a printer 503 on a dot pattern paper 301 on which a dot pattern is printed. Thereby, a form 100 is created (form creation). The person who performs overprinting does not necessarily need to be a printing company. For example, the user may create the form 100 by overprinting on the dot pattern paper 301.
The technique of the present invention can also be used regardless of the type and presence of the dot pattern.

The user fills in the necessary items using the electronic pen 101 for the form 100 created by overprinting. The entry data is stored as trajectory information in a storage unit (not shown) built in the electronic pen 101 (storage of trajectory information), and is appropriately transmitted to a personal computer (hereinafter referred to as “PC”) 102 by a user ( Trajectory information capture).
The trajectory information transmitted to the PC 102 is transmitted to the server (processing device) 104 via the communication line 103 (trajectory information transmission).
The server 104 performs identification processing for identifying the form type corresponding to the trajectory information using the transmitted trajectory information (form identification processing).

Subsequently, the server 104 transmits the identified form type information and the trajectory information to the PC 105 (transmission of the form type and the trajectory information).
The PC 105 displays the received form type information and trajectory information on a screen or the like (screen display). In this way, support is provided when the determiner determines the form type.

  Next, with reference to FIG. 1, the configuration of each component and server 104 used in the system will be described with reference to FIG. FIG. 2 is a block diagram showing the components of FIG. 1 and the functions of the server. In addition, the same code | symbol is attached | subjected to the same element as FIG.

  The system includes an electronic pen 101 for filling in a form 100 and acquiring trajectory information (entry data), a PC 102 for acquiring the trajectory information from the electronic pen 101 and transmitting it to the server 104, and the trajectory information from the PC 102 to the server 104. The communication line 103 for transmitting to the server, the server 104 for performing the process of identifying and selecting the type of the form 100 from the transmitted locus information and the data stored in itself, and the form 100 identified by the server 104 It is mainly composed of the PC 105 that displays the type and the like.

(Electronic pen)
The electronic pen 101 used in the present embodiment transmits the entry data entered in the form 100 by the user to the PC 102 as trajectory information.
More specifically, the configuration of the electronic pen 101 acts on the electronic pen 101 at the time of entry, for example, a camera for reading a dot pattern printed on the form 100, although not shown, in addition to a normal ink-type pen tip. A pressure sensor that detects writing pressure, a processor that processes data from the camera and the pressure sensor, associates the position of the pen tip moved by entry with the dot pattern, creates it as locus information, stores the created locus information, etc. A storage unit, a communication unit, and the like are mounted.

  Examples of the communication unit include a device that transmits and receives information using radio waves in a predetermined frequency band according to a communication protocol such as Bluetooth (registered trademark) that is a kind of short-range wireless transmission technology. A device that transmits and receives information according to a communication protocol of a wireless LAN (Local Area Network) or wired communication technology may be used. The processor is composed of electric / electronic circuits such as a CPU (Central Processing Unit) and a ROM (Read Only Memory), and controls the electronic pen 101 in an integrated manner.

Specifically, the electronic pen 101 detects the pressure applied to the pen tip when the user writes a character or the like on the form 100 with the electronic pen 101, that is, the writing pressure, and supplies the writing pressure data to the processor. The processor switches the camera on and off based on the writing pressure data provided from the pressure sensor. The electronic pen 101 determines that the user has started entry when a writing pressure of a predetermined value or more is detected, and activates the camera. The camera reads the dot pattern on the form 100 and supplies the information to the processor. The processor calculates the X / Y coordinates of the electronic pen 101 on the form 100 from the supplied information.
The processor associates the calculated X / Y coordinates with the time stamp (time information) and saves the stored trajectory information in the storage unit as time-series trajectory information, and sends the saved trajectory information to the PC 102 via the communication unit according to a user instruction. To send.
Note that the technology of the present invention is not limited to an electronic pen as long as it is a device that can store trajectory information entered by a user.

(Form)
The form 100 filled with the electronic pen 101 functions as a dedicated paper for the electronic pen by using, for example, a paper on which a special dot pattern is printed. In this dedicated paper, dots are arranged so that the position can be specified when the paper surface is divided into grids of a predetermined size, and the pen tip of the electronic pen 101 on the form 100 is used using the information of this dot pattern. The position of can be specified.
In the printing company in FIG. 1, a form 100 is generated by printing a format such as ruled lines, frames, item names, and the like, on the dedicated paper. This format difference is the form type in the present invention.
The user of the electronic pen 101 who fills in the form 100 fills in the form 100 by specifying the entry position and the like from the format (ruled lines, frames, item names, etc.) printed on the upper surface of the form 100.

  Usually, when the form type is different, the dot pattern paper 301 (FIG. 1) having a different dot pattern is used so that the system can identify the form type. However, the cost is often high due to the license fee and the like. . Some dot patterns have a high degree of freedom in layout, such as those that do not have format specifications. Therefore, in the present invention, dot pattern paper 301 on which the same dot pattern is printed is used even if the form types are different, such as form A, form B, and form C. Then, the form type is identified based on the difference in the writing position due to the difference in the format of the form 100 (layout such as ruled lines and frames).

FIG. 3 shows an example of each form generated by overprinting the format on the dot pattern paper.
On the dot pattern paper 301, the form A is an overprint of the format 400A, the form B is an overprint of the format 400B, and the form C is an overprint of the format 400C.
Hereinafter, reference numeral 100 is used to indicate an arbitrary form, and reference numeral 400 is used to indicate an arbitrary format.

FIG. 4 is a diagram showing a specific example of each form and an example of entry with the electronic pen.
The form A is a form 100 having a receipt format layout for entering the company name and amount. An example of entry data on the form A is shown in an entry example 410A.
The form B has an order table layout in which the company name, product name, quantity, amount, and total number are entered. An example of entry data in the form B is shown in an entry example 410B.
The form C is similar to the layout of the form B, but a code 4001 for entering a code corresponding to the product name is added. An example of entry data on the form C is shown in an entry example 410C.
Thus, the forms (A, B, C) have different formats (410A, 410B, 410C) depending on their types, and the character entry positions are different.

(Personal computer)
Returning to FIG.
Although not shown, the PC 102 generally includes at least a storage unit as a storage unit used when performing arithmetic processing, an arithmetic processing unit that performs the arithmetic processing, and a network interface used when performing communication. Configured as a simple computer. The storage unit is configured by a RAM or the like. The arithmetic processing is realized by an arithmetic processing unit configured by a CPU executing a program on the storage unit.
The PC 102 receives the trajectory information transmitted via the communication unit built in the electronic pen 101, and saves it in the storage unit built in itself as necessary.
Then, the stored trajectory information is transmitted to the server 104 via the network interface and the communication line 103 at a predetermined timing. This timing may be instructed by the user or may be periodically performed.

(server)
The server 104 receives the trajectory information transmitted from the PC 102, performs an operation processing unit 110 that performs processing such as identification of a form type, a RAM 111 that temporarily stores data during the processing of the operation processing unit 110, and operation processing It is mainly composed of a transmission / reception unit 112 that transmits a processing result and the like by the unit 110 to the PC 105, a storage unit 120 as a storage unit, and the like. In the communication between the PC 102 or the PC 105 and the server 104, the description is omitted below through the transmission / reception unit 112.

  The storage unit 120 of the server 104 stores an identification program (hereinafter referred to as “program”) 121. The program 121 is expanded in the RAM 111 of the server 104 and executed by the arithmetic processing unit 110, thereby functioning as an identification unit.

Further, the storage unit 120 stores a trajectory database (hereinafter referred to as “DB”) 125 that stores trajectory information of the electronic pen 101, and a form (A, B, C, see FIG. 3) for identifying the form type. A form area definition master (hereinafter referred to as “area definition master”) 126 for storing area definitions and a related DB 127 for storing related data are stored.
Note that the trajectory DB 125, the area definition master 126, and the related DB 127 are configured in the storage unit 120 of the server 104 in FIG. 2, but all or any one of them is provided outside the server 104 and has a predetermined transmission. It may be connected by a line or the like.

The trajectory DB 125 holds a trajectory information table 500 shown in FIG. In the trajectory information table 500, one record represents an X coordinate and a Y coordinate in the form 100 of the electronic pen 101 for each hour, an ID 5001 that is a unique ID (IDentification) given to each record, and a time representing time 5002, fields of an X coordinate 5003 and a Y coordinate 5004 in the form 100 are included. For example, the record in the row 5010 indicates that the pen point of the electronic pen 101 is X coordinate 5003 = “9” at the time “2005/11/01 11: 09: 00.01” at time 5002 for the data with ID 5001 = “1”. Y coordinate 5004 = “4”.
In addition to the above items, for example, information such as a unique ID is further added for each sheet of the form 100, and managed by dividing from the start (start of entry) to the end (fill in the check mark of entry end). May be.

  The area definition master 126 holds an area definition table 600 shown in FIG. In the area definition table 600, one record represents the number of points for each area of each form (A, B, C), form name 6001, area ID 6002, start X coordinate 6003, end X coordinate 6004, start Y coordinate. 6005, an end Y coordinate 6006, and a field of a given point 6007 representing a point to be added when trajectory information exists. The area definition table 600 corresponds to entry position definition information in claim 1.

The area division in FIG. 6B schematically represents an example of the area division for a certain form 100. Here, as an example, the form 100 is divided into 30 parts, IDs of 01 to 30 are assigned to respective areas in which the X coordinate is taken on the horizontal axis, the Y coordinate is taken on the vertical axis, and the respective coordinates are divided into 10 parts. For the divided areas, area IDs are assigned in ascending order in order from the smaller coordinate (upper left of the form 100).
Based on this area ID, an area ID 6002, a start X coordinate 6003, an end X coordinate 6004, a start Y coordinate 6005, and an end Y coordinate 6006 in FIG. 6A are set. That is, the range defined by these four values is the range of the corresponding area.
For example, a row 6010 in FIG. 6A shows a start X coordinate 6003 = “0” and an end X coordinate 6004 = “area corresponding to area ID 6002 =“ area 01 ”of the form name 6001 =“ form A ”. 10 ”, start Y coordinate 6005 =“ 0 ”, and end Y coordinate 6006 =“ 10 ”. If trajectory information exists in the area in “area 01”, 5 points (hereinafter, points are appropriately described as “P”) are added as the form A from the grant point 6007. A row 6011 indicates that 5P is added to the form B when the trajectory information exists in the same area as the “area 01” in the row 6010. The point addition will be described later with reference to FIG.
In the present embodiment, this area division is common to each form (A, B, C), but each area may be defined by a different division.

FIG. 7 is a diagram schematically showing, as an example, the value of points given for each area in each form. 7A shows the grant point definition for each area of the form A, FIG. 7B shows the grant point definition for each area of the form B, and FIG. 7C shows the grant points for each area of the form C. Definition.
For example, in the definition of granted points for each form A area in FIG. 7A, reference numeral 7010 corresponds to the information in the row 6010 in FIG. 6A, and “trajectory information” exists in “area 01”. It is schematically shown that “5P” is given. The same applies to the other areas, and the same applies to the definition of points given for each form B area in FIG. 7B and the definition of points given for each form C area in FIG.

Returning to FIG.
The related DB 127 stores related information acquired by the electronic pen 101. The related information is, for example, information on the form identification processing result, information on the total number of points for each form described later with reference to FIG. 9, the character type master in FIG. 11, and the like, but may be deleted if unnecessary.

  The server 104 uses the program 121, the trajectory DB 125, the area definition master 126, the related DB 127, and the like to transmit information such as the identified form type candidates to the PC 105.

(Personal computer)
The PC 105 is a general computer similar to the PC 102, and is configured to include at least an arithmetic processing unit (not shown), a network interface, a display unit, and an input unit.
The PC 105 receives the information transmitted from the server 104 via the network interface and displays it on the display unit or the like. The determination person visually confirms the information of the displayed form type, and determines the form type corresponding to the trajectory information.

(Processing of the entire identification system)
FIG. 8 is a flow of processing of the identification system in the present embodiment.
Detailed processing and operation of the present system will be described with reference to FIG. 8 and with reference to FIGS. In addition, regarding each process (S201-S207) in FIG. 8, the same code | symbol is attached | subjected about the same process as FIG.
In this embodiment, a case where the electronic pen 101 is used to fill in the form B will be described as an example.

  First, for example, a printing company generates a form B by overprinting the format 400B on the dot pattern paper 301 (S201).

  Subsequently, entry data written by the user with the electronic pen 101 on the generated form B is stored as trajectory information by the storage unit of the electronic pen 101 (S202). The trajectory information stored here is trajectory information including X / Y coordinates of the time unit of the electronic pen 101 as described above.

  Next, the trajectory information stored in the electronic pen 101 is captured by the PC 102 using a predetermined communication path (S203).

  Subsequently, the PC 102 transmits the captured locus information to the server 104 via the communication line 103 (S204). The server 104 stores the trajectory information received from the PC 102 as the trajectory information table 500 in the trajectory DB 125 of its own storage unit 120.

  The arithmetic processing unit 110 of the server 104 uses the stored trajectory DB 121 and the area definition master 126 to identify in which form (A, B, C) the received trajectory information is entered ( (Form identification processing) (S205). Details of the form identification process will be described later with reference to FIG.

  The form type information identified by the server 104 is transmitted to the PC 105 as a form type candidate, triggered by a request from the PC 105 or the like. At that time, the trajectory information stored in the trajectory DB 125 is also transmitted (S206). As for the timing of transmission to the PC 105, the form type information may be stored in the related DB 127 and read and transmitted from the related DB 127 each time the PC 105 requests, or the information of the related DB 127 may be sent periodically. May be transmitted to the PC 105.

  The arithmetic processing unit of the PC 105 displays the form type candidate information and the trajectory information received from the server 104 on the screen of the PC 105 (S207). An example of the screen displayed here will be described later with reference to FIGS.

(Form identification process flow)
FIG. 9 is a detailed process flow of the form identification process (S205) in FIG. A description will be given along FIG. 9 with reference to FIGS. As described above, this form identification process functions when the program 121 is expanded in the RAM 111 of the server 104 and executed by the arithmetic processing unit 110. The description is omitted below.
Here, it is assumed that information on the total number of points for each form is stored in advance in the related DB 127 in FIG. Detailed explanation will be given later.

First, the arithmetic processing unit 110 reads the trajectory information table 500 from the trajectory DB 125 of the storage unit 120 into the RAM 111 or the like (S2051).
Subsequently, the arithmetic processing unit 110 sets 1 as an initial value in the loop counter n (S2052).

The arithmetic processing unit 110 reads the area definition table 600 of the nth form 100 from the area definition master 126 of the storage unit 120 into the RAM 111 or the like (S2053).
For example, when n = 1 (form A), data of form name 6001 = “form A” is read in the area definition table 600.
Using the trajectory information table 500 read in S2051 and the area definition table 600 read in S2053, the total number of points of the nth form 100 is calculated (S2054). Hereinafter, a method for calculating the total number of points will be described.

The arithmetic processing unit 110 uses the information in the trajectory information table 500 and the area definition table 600 read into the RAM 111 and the like, and adds points for each area of the area definition table 600 that matches the trajectory information table 500 for each form. . That is, for each area in the area definition table 600, it is determined whether or not the trajectory information exists in the trajectory information table 500. If the trajectory information exists, the grant points set in the area are displayed as a form. It is to add every time.
For example, locus information (entry data) exists at <X coordinate “9”, Y coordinate “4”> in the row 5010 of FIG. 5. In the area definition table 600 of FIG. ”, Y coordinate“ 4 ”> in line 6010 having in the area, the grant point 6007 =“ 5P ”is added to the form A. If trajectory information exists in other areas, points are added in the same manner based on the information on the granted points 6007 in the area definition table 600, and the total number of points is calculated.

FIG. 10A schematically shows the trajectory information table read in S2051, and is a diagram superimposed on the area divisions of the area definition table.
FIG. 10B is a diagram schematically showing the grant points of each area for the form A. Here, the area with the trajectory information is shaded. With reference numeral 1011, “area 01” is given “5P” because the trajectory information exists (shaded display) from the reference numeral 1001 (FIG. 5, line 5010) in FIG. Further, “area 05” has locus information (shaded display) as shown in FIG. 10A, but no points are assigned to “area 05” in the form A because no points are assigned to “area 05”. “Area 19” is not given points because there is no trajectory information. The same applies to other areas.
As a result, the total number of points in the form A of the trajectory information is 11P (shaded display part: 5P + 5P + 1P).

Subsequently, the arithmetic processing unit 110 reads information on the total number of points of the nth form 100 stored in advance in the related DB 127 of the storage unit 120 into the RAM 111 or the like (S2055). The information on the total number of points of the form 100 is obtained by adding the granted points for each form (A, B, C) defined in the area definition table 600. For example, the total number of points of the form A is shown in FIG. 7 to 12P (5P + 5P + 1P + 1P), similarly, information that the form B is 47P and the form C is 51P is stored in the related DB 127 in advance.
Here, an example has been described in which the total number of points of each form (A, B, C) is stored in the related DB 127 in advance. However, the present invention is not limited to this. Each time the form identification process is executed, the area definition is defined each time. The total number of points of each form (A, B, C) may be calculated using the table 600.

Subsequently, the matching rate of the total number of points with respect to the total number of points in the nth form 100 is calculated (S2056). The relevance rate corresponds to the degree of coincidence in claim 1 and is obtained by the sum of points / total points. In addition, it means that the probability that it is a candidate is high, so that the number of all points of each form (A, B, C) and the total number of points are close (high relevance rate). In this case, the total number of points in the trajectory information does not exceed the total number of points in the form 100.
When the relevance rate is calculated by the above-described method, the form A has a relevance rate of 0.9166 (91.7%) with the total number of points 11P out of the total number of points 12P.

  The arithmetic processing unit 110 determines whether or not the value of the loop counter n is equal to or greater than the number of types in the form name 6001 of the area definition table 600 read into the RAM 111 or the like in S2053 (n ≧ number of types of form) (S2057). ).

  When n <the number of types of forms (S2057 → No), the arithmetic processing unit 110 increments n (S2058) and returns to the processing of S2053. As a result, the precision of the next form 100 is calculated.

For the forms (A, B, C), examples of the result of calculating the total number of points in S2054 are schematically shown in FIGS.
As described above, FIG. 10B is a diagram schematically showing the grant points of each area for the form A. Similarly, FIG. 10C schematically shows the grant points of each area for the form B, and FIG. 10D schematically shows the grant points of each area for the form C.

For the form (A, B, C), the result of calculating the relevance rate in S2056 is as follows.
Form A has an accuracy rate of 0.9166 (91.7%) with a total of 11 points out of a total of 12 points.
Form B has a precision of 1.000 (100.0%) with a total of 47 points out of 47 points.
Form C has a precision rate of 0.9215 (92.2%) with a total of 47 points out of 51 points.
Based on the above calculation results, candidates of form B, form C, and form A are identified in descending order of the relevance ratio as the form type in which the locus information is described.

  On the other hand, if n ≧ the number of forms (S2057 → Yes), the form identification process (S205) ends, and the process proceeds to S206 in FIG. That is, in S206 of FIG. 8, the server 104 transmits the information on the form B, the form A, the form C, and the information on the trajectory DB 125 to the PC 105 based on the precision calculated in S205. .

  In addition, as a modified example of the precision calculation, each area (A, B, C) can be set in advance by setting different area classifications for each form (A, B, C) or by setting different points for each area. , B, C) may be the same (for example, 100P). In that case, the calculation of the relevance ratio can be omitted by directly comparing the total points of each form (A, B, C).

In addition, as an example of calculating the total number of points, the method of adding grant points has been described. However, as a modification, for example, a method of subtracting (penalty method) may be used. Specifically, a negative value (such as “−2”) is given as a point for an area where the trajectory information should not exist (for example, areas 03, 04,... In FIG. 10B). It is also possible to compare the sum of negative values.
Further, by combining the subtracting method and the above-described adding method, it is possible to widen the difference in the total number of points for each form and facilitate comparison.

Furthermore, as a modification of the point calculation, for example, the character type of the trajectory information may be determined, and the point may be calculated for each determined character type.
FIG. 11 shows an example of the character type master stored in the relation DB 127 of FIG. The character type master 1100 has a form name 1101, an area ID 1102, a character type 1103, and a grant point 1104. The form name 1101 is linked to the form name 6001 of the area definition table 600, and the area ID 1102 is linked to the area ID 6002 of the area definition table 600. The character type of the trajectory information is set for each area of each form (A, B, C). The points can be calculated by discrimination.
The point calculation method is as follows. In reference numeral 1001 in FIG. 10A, it is determined that there is trajectory information in “area 01”, and the trajectory information is used by using an existing character identification technique such as Japanese Patent Application Laid-Open No. 2003-308483. It is determined whether (entry data) is kanji or kana. In the row 1110 of the character type master 1100, if the trajectory information of “area 01” is “kanji or kana”, the grant point 1104 = “5P” is assigned, so that as indicated by reference numeral 1011 in FIG. 5P is given as a point of “area 01” of the form A. If the trajectory information of “area 01” is “numeric”, 1P is assigned from row 1111. The same applies to the forms B and C.
When the character type master 1100 is used, the value of the grant point 1104 in FIG. 11 is used for point calculation, so the item of the grant point 6007 in FIG. 6 is unnecessary.
In this modification, points can be defined by area, character type, and the like.

(Operation screen)
12 to 14 are operation screens for supporting the determination of the form type by the determiner. By determining and confirming the form type on the operation screen 1201, the form type corresponding to the trajectory information is confirmed.

FIG. 12 is a screen example when “Form B” is selected in the form candidate area of the operation screen.
In this form candidate area 1208, the candidates for the form (A, B, C) identified by the process of S205 are displayed in a list in the form name display area 1210 in descending order of the relevance ratio. In this case, the form B having the highest relevance ratio among the forms (A, B, C) is selected in advance, and ◯ indicating that is displayed in the selection area 1209.

  In the locus information display area 1205, information obtained by superimposing locus information acquired from the electronic pen 101 on the format 400B (FIG. 3) of “form B” selected in the selection area 1209 of the form candidate area 1208 is displayed. Yes. The determiner can visually check the trajectory information display area 1205 and determine whether or not the form type selected in the selection area 1209 matches the trajectory information.

  As a result of determination by the determiner, if the form type is in the trajectory information, the determination information is selected in the selection area 1209 by pressing the OK button 1211 for confirming the form type via the input unit. It is determined as corresponding to the form type. The information may be stored in the related DB 127 of the storage unit 120 of the server 104, for example.

  When the determiner presses the cancel button 1212 via the input unit, the form type regarding the trajectory information is not fixed. Here, if the form type has never been confirmed, the form type is undetermined.

  In the example of FIG. 12, both the display frame 1206 and the display frame 1207 match the trajectory information, and it can be understood that the form B is a form type corresponding to the trajectory information (entry data). When the result of the determination is viewed by the determination person, the determination person presses the OK button 1211 via the input unit, and confirms that the trajectory information displayed in the trajectory information display area 1205 is information on the form B.

  FIG. 13 is an example of a screen when the determiner selects “form C” in the form candidate area of FIG. In the selection area 1209 of the form candidate area 1208, “◯” indicating that “form C” is selected is displayed.

The locus information display area 1205 displays information obtained by superimposing locus information acquired from the electronic pen 101 on the format 400C (FIG. 3) of the form C.
In the display frame 1302 of the trajectory information display area 1205, the row of codes 1303 does not match the trajectory information. From this result, when the determiner determines that the form C is not appropriate for the trajectory information, when the cancel button 1212 is pressed, the form type of the trajectory information is not changed.

(Modification)
FIG. 14 is an example of a screen on which the image previously captured with an electronic pen is added and displayed on the operation screen of FIG. 12 with a scanner or a camera.
In this case, for example, the same form ID information is assigned to the locus information and the captured image, and the server 104 (FIG. 1) links the assigned form IDs to the locus information. The corresponding image is displayed on the captured image 1401. As a result, the determiner can determine the form type while comparing the image image with the trajectory information. Note that it is possible to further ensure the determination of the form type by displaying the captured image 1401 on the entire operation screen 1201 or displaying the captured image 1401 by enlarging / reducing or the like.
This modification is effective when the judge cannot confirm the actual form 100 actually entered, or when the memory of the judge 100 is ambiguous.

  In the present embodiment, the candidate form types are displayed as a list in order of the relevance ratio. However, the present invention is not limited to this, and only the form type having the highest relevance ratio may be displayed. In the present embodiment, an example in which three candidate form types are displayed as a list has been described. However, the present invention is not limited to this, and a predetermined value is set in advance for a matching rate, and a matching rate equal to or higher than that value is acquired. All candidates may be displayed in a list.

Furthermore, in the present exemplary embodiment, the method in which the server 104 executes the form identification process by acquiring the trajectory information from the electronic pen 101 by the PC 102 and transmitting it to the server 104 has been described as an example. Then, the form identification process may be executed using the trajectory information acquired from the electronic pen, and the determiner may determine the form type in the PC 102 based on the result. In that case, the server 104 and the PC 105 are unnecessary. Alternatively, the trajectory information may be transmitted from the PC 102 to the PC 105 without going through the server 104, the PC 105 may execute a form identification process, and the determiner may determine the form type based on the result.
Other modifications can be made without departing from the spirit of the present invention.

(Other)
According to the embodiment described above, even if a plurality of form types are used for one dot pattern, the form type is determined using the trace information by the electronic pen, and the trace information is matched to the form type. By displaying the information, information is not added to the form in advance, and the user does not need to perform extra operations when filling in the form.
The license per dot pattern may incur costs, and the use of different dot patterns to identify forms has been required to improve in terms of manufacturing and cost. However, according to the present embodiment, it is not necessary to prepare a dot pattern for each form.
In addition, the technique of the present embodiment can be used in the trajectory information processing technique that does not use a dot pattern.
In addition, each form is displayed in a list in order of candidates, and by selecting a form type that is a candidate on the operation screen, the trajectory information is superimposed on each form type, and the form type of the determiner is displayed. Confirmation work can be supported.

  An electronic pen system based on Anoto (registered trademark) defines one form for each dot pattern. This system can support the definition of a plurality of form types (formats) within the scope of the form definition, and can reduce the cost associated with the license acquisition of the dot pattern. Further, in the case where there is no restriction on the form definition as described above, it is possible to design a free form type (layout) individually, so there is an advantage that the restriction is reduced and the design of the form is facilitated.

  This system uses the electronic pen's trajectory information to determine the form type, etc., based on the predefined number of points for each area and the total number of points for each character type (type of kanji, numbers, English, etc.). It is possible to identify. Therefore, the present system can be applied not only for identification of the form type and the like, but also for the whole classification based on identification of user entry contents (trajectory information). For example, it becomes easy to select a paper whose document format (layout, character type, etc.) is determined, and to identify accounting information, personnel information, and the like. Further, the present invention can be widely applied as a data format or form format identification function by analysis of data type and data arrangement.

It is a block diagram with which the identification system regarding one Embodiment of this invention is applied. FIG. 2 is a block diagram showing functions of the components of FIG. 1 and a server. It is an example of each form produced | generated by overlapping and printing a format on the dot pattern regarding one Embodiment of this invention. It is the figure which showed the specific example of each form regarding one Embodiment of this invention, and each example of entry with an electronic pen. It is a figure which shows the example of the locus | trajectory information table stored in locus | trajectory DB regarding one Embodiment of this invention. (A) It is the figure which represented typically the example of the area definition table stored in the area definition master, (b) The example of the area division of a form regarding one Embodiment of this invention. The value of the point given for each area of each form relating to one embodiment of the present invention is defined as (a) point definition per form A area, (b) point definition per form B area, (c) point definition per form C area, It is the figure typically shown as. It is a processing flow of the identification system according to an embodiment of the present invention. FIG. 9 is a detailed process flow of the form identification process of FIG. 8. FIG. (A) A diagram schematically showing the trajectory information table read in S2051 and superimposed on the area division of the area definition table, (b) a diagram schematically showing each area grant point of the form A (C) It is the figure which showed typically each area | region grant point of the form B, and (d) The figure which showed typically each area grant point of the form C. It is a figure which shows the example of the character type master preserve | saved in related DB regarding one Embodiment of this invention. It is a figure which shows the example of a screen when the form B is selected in the form candidate area of an operation screen. It is a figure which shows the example of a screen when the form C is selected in the form candidate area of FIG. FIG. 13 is a diagram showing an example of a screen on which a form preliminarily filled in with an electronic pen is added and displayed on the operation screen of FIG.

Explanation of symbols

100 form 101 electronic pen 102,105 personal computer 104 server (processing device)
1201 Operation screen 121 Identification program 125 Trajectory database (trajectory DB)
126 Form area definition master (area definition master)
1401 Captured image 301 Dot pattern paper 400, 400A, 400B, 400C Format 500 Trajectory information table 600 Area definition table (entry position definition information)
6007 points awarded

Claims (7)

A form created using the prescribed paper,
An electronic pen that generates trajectory information by writing to the form;
In the identification system for identifying the type of the form using a processing device that analyzes the trajectory information acquired from the electronic pen and converts it into electronic data,
The processor is
Storage means for storing entry position definition information defined due to different entry positions by the electronic pen for each form type;
Using the entry position definition information and the locus information input from the electronic pen by writing to the form to be identified, the degree of coincidence between the entry position definition information and the locus information is calculated, An identification means for identifying the calculated high-matching type as a candidate,
An identification system comprising: an output unit that allows the determination person to select the type of the form by displaying the identified type information and the trajectory information on a screen. A form created using the prescribed paper,
An electronic pen that generates trajectory information by writing to the form;
In the identification system for identifying the type of the form using a processing device that analyzes the locus information acquired from the electronic pen and converts it into electronic data,
The processor is
Storage means for storing entry position definition information defined due to different entry positions by the electronic pen for each form type;
Using the entry position definition information and the locus information input from the electronic pen by writing to the form to be identified, the degree of coincidence between the entry position definition information and the locus information is calculated, An identification system comprising: identification means for identifying a type having the highest degree of matching among the calculated types. The entry position definition information is:
The form is divided into predetermined areas, and includes information that gives predetermined points according to the presence or absence of trajectory information for each area,
The identification means includes
It is determined whether or not the trajectory information input from the electronic pen exists for each area, and the total of the predetermined points given to the area is calculated according to the presence or absence, and the calculated points The identification system according to claim 1, wherein the degree of coincidence is calculated by comparing the number and the total number of points calculated for each type of form. To make the form type selectable,
The identification means includes
Identify the types of forms with a high degree of coincidence with the trajectory information as a plurality of candidates in order of the degree of coincidence,
The output means includes
The identification system according to claim 3, wherein the output is performed in descending order of the degree of coincidence. The identification means includes
The identification system according to claim 4, wherein a type of form having a degree of matching equal to or greater than a predetermined value is identified as a candidate. The processor is
The image information of the form filled in with the electronic pen is acquired, and the image information is displayed on the screen in addition to the information of the identified type and the trajectory information by the output means. 6. The identification system according to any one of items 5. A form created using the prescribed paper,
An electronic pen that generates trajectory information by writing to the form;
In the identification system for identifying the type of the form using a processing device that analyzes the locus information acquired from the electronic pen and converts it into electronic data,
Storage means for storing entry position definition information defined due to different entry positions by the electronic pen for each form type;
Using the entry position definition information and the locus information input from the electronic pen by writing to the form to be identified, the degree of coincidence between the entry position definition information and the locus information is calculated, An identification means for identifying the calculated high-matching type as a candidate,
An identification program that causes a processing device, which is a computer, to execute an output unit that displays the information of the identified type and the trajectory information on a screen so that a determiner can select the type of the form.

Images