JPH08221508A - Character recognition processor - Google Patents

Abstract

PURPOSE: To automatically recognize fields to be subjected to a link knowledge processing and to link them. CONSTITUTION: A candidate character string 9 is obtained by reading characters, which are described on a form 1 having plural fields, and performs character recognition processing. An automatic Japanese syllabary (KANA)/Chinese character (KANJI) link processing part 3A retrieves a handwritten KANA field and a handwritten KANJI field, whose positions of the form 1 are in the previously decided relation, and links these fields. For the respective linked fields in the candidate character string 9, a link knowledge processing part 5 performs the link knowledge processing. For the other fields, a single knowledge processing part 4 performs a single knowledge processing.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a character recognition processing apparatus, and more particularly, to a character recognition processing apparatus for automatically cooperating with each other in the character recognition processing of kana and kanji written on a form to perform knowledge processing. .

[0002]

2. Description of the Related Art FIG. 6 is a background illustration of the present invention.
An example of a general form used in a character recognition processing device such as an R (optical character reading device) is shown.

In the form 1 in FIG. 6, the member number,
Application date, zip code, telephone number, address (kana), address (kanji), name (kana), name (kanji), date of birth, occupation,
The age and gender are entered in predesignated areas (fields).

Of these, only the member number is a type field in which type characters are pre-filled, and the other is a handwriting field in which handwriting characters are written by the person who fills out the form. Of the handwritten fields, the date of application, zip code, telephone number, date of birth (the latter half), occupation, and age are numerical fields filled in with numbers, and the date of birth (the first half),
Gender is a mark field that is filled in by a method of putting a mark (circle mark) on a previously written type character, and each address kana, name kana field is a kana field written in kana characters, address kanji, Each field of the name and kanji is a kanji field written in kanji.

The character recognition processing device reads characters entered in a form as shown in FIG. 6 and performs character recognition processing. In this character recognition process, in order to improve the character recognition rate (the rate at which the written characters are correctly read), for addresses and names, the character strings that were once read and recognized from the Kana and Kanji fields (candidate character strings) ), A predetermined knowledge process (independent knowledge process) is performed. This independent knowledge process is performed for each field and is a process of correcting a candidate character string using an address dictionary and a name dictionary prepared in advance. For example, the candidate character string read and recognized from the address kana field is collated with the address kana dictionary, and even if the candidate character string contains an error of 2 to 3 characters, this is corrected and output as a correct character string. To do.

Further, as one of the knowledge processes, a kana field and kanji field cooperation process (cooperation knowledge process) for an address and a name are performed. This cooperative knowledge process is a knowledge process in which not only the knowledge process is independently performed for each candidate character string read from the kana field and the kanji field, but also two fields designated in advance are cooperated.

In the above-mentioned form 1, for example, "address"
There are two fields for entering "Kana" and "Kanji". Therefore, for example, the candidate character string obtained from the address Kanji field is corrected using the candidate character string obtained from the address Kana field, and / or the candidate character string obtained from the address Kana field is obtained using the candidate character string obtained from the address Kanji field. Correct the candidate character string. Thereby, the character recognition rate in the character recognition processing can be increased.

[0008]

According to the above-mentioned conventional technique, the kana field and the kanji field, which are relatively difficult to recognize because of the large number of characters to be entered and the handwriting, are processed by cooperative knowledge processing. The character recognition rate can be improved.

However, in the above-mentioned conventional technique, it is necessary for the operator to previously specify the kana field and the kanji field for which the cooperative knowledge process should be performed in the recognition information definition process.

For example, when character recognition processing is performed on the form shown in FIG. 6, there are two portions where cooperative knowledge processing should be performed. That is, an address kana field and an address kanji field, and a name kana field and a name kanji field. In this case, prior to the character recognition processing, the operator inputs an instruction to link the address kana field and the address kanji field in the recognition information definition processing,
Input to instruct to link the name kana field and the name kanji field.

The input of the instruction for the cooperative knowledge process is performed as follows in the recognition information definition process. For example, the operator designates each field as an area to be recognized (recognition field), thereafter inputs the attribute of each field, and gives an item number to each field. Then, for a certain field, if there is a field to be linked with this, the field to be linked is specified using the item number, and then an instruction to link these is input.

As described above, according to the above-mentioned conventional technique, the operator needs to make an input for instructing the fields to be linked as many times as the number of portions of the form for which the linking knowledge processing should be performed. Therefore, as the number of parts to be subjected to cooperative knowledge processing increases, the burden on the operator increases accordingly. Further, it is necessary to input this instruction for each type of form 1, which is a burden on the operator. Further, since the number of times the operator operates the character recognition processing device is large, the character recognition processing device itself has been accelerated, but the efficiency of the character recognition processing cannot be improved. Further, since the input is made by the operator, there is a possibility that an error in inputting the instruction of the cooperative knowledge processing may occur.

It is an object of the present invention to provide a character recognition processing device that automatically recognizes fields for which cooperative knowledge processing should be performed and links them. Also, the present invention
An object of the present invention is to provide a character recognition processing device that automatically recognizes a handwritten kana field and a handwritten kanji field for which cooperative knowledge processing should be performed and links them.

[0014]

FIG. 1 is a principle block diagram of the present invention, showing a configuration of a character recognition processing apparatus according to the present invention. This character recognition processing device reads characters entered in each field of a form 1 having a plurality of predetermined fields, performs character recognition processing for each field, and obtains a character recognition output 6 Is.

This character recognition processing device comprises an independent knowledge processing unit 4, a cooperative knowledge processing unit 5, and an automatic cooperation processing unit 3.
The independent knowledge processing unit 4 performs the knowledge process in the character recognition process for a predetermined field by itself. The cooperative knowledge processing unit 5 performs the knowledge process in the character recognition process for a predetermined field in cooperation with the field and other fields associated with the field. The automatic cooperation processing unit 3 selects from among a plurality of fields
A plurality of fields whose positions on the form 1 have a predetermined relationship are searched and linked. At this time, in particular, the automatic cooperation processing unit 3 cooperates the handwritten kana field and the handwritten kanji field. For a plurality of fields linked by the automatic link processing unit 3, the link knowledge processing unit 5 links these to perform the knowledge processing.

[0016]

In most cases, the kana field and the kanji field to be linked in the knowledge processing in the character recognition process, especially the handwritten kana field and the handwritten kanji field, have a predetermined relationship on the form 1. For example, they are provided at extremely close positions or adjacent positions. Therefore, using this, the automatic cooperation processing unit 3
Automatically retrieves a plurality of fields having a predetermined relationship on the form 1, particularly a handwritten kana field and a handwritten kanji field, and links them.

Therefore, according to the character recognition processing apparatus of the present invention, it is possible to automatically associate a plurality of fields in correspondence on the form 1, particularly a handwritten kana field and a handwritten kanji field. This allows
In most of the forms 1, a plurality of predetermined fields, in particular, a handwritten kana field and a handwritten kanji field can be made to cooperate with each other, and cooperative knowledge processing can be performed for these fields.

As a result, it is possible to eliminate the need to specify the kana field and the kanji field for which the cooperative knowledge process should be performed in the recognition information definition process. As a result, when performing character recognition processing, the operator does not need to input instructions for cooperative knowledge processing, the burden on the operator can be extremely reduced, and errors in instruction input for cooperative knowledge processing can be prevented. it can. Further, even when the types of the form 1 are different, it is possible to eliminate the need to input the instruction of the cooperative knowledge processing each time. Moreover, since the number of operations by the operator can be reduced, the efficiency of character recognition processing can be improved.

[0019]

FIG. 2 is a block diagram of an embodiment, showing a character recognition processing device. The character recognition processing apparatus according to the present embodiment reads a character written on the form 1, a reading processing unit 7, performs a character recognition process on a result read by the reading processing unit 7, and a candidate character string 9
And a knowledge processing unit 11 that obtains a character recognition output 6 by performing knowledge processing on the candidate character string 9. The reading processing unit 7, the character recognition processing unit 8, and the knowledge processing unit 11 are provided in the processing device 2. The processing device 2 includes a CPU (central processing unit) of the character recognition processing device,
Each processing program on the memory of the character recognition processing device. The processing programs include, for example, a reading processing program forming the reading processing unit 7, a character recognition processing program forming the character recognition processing unit 8, a knowledge processing program forming the knowledge processing unit 11, and the like.

When the form 1 is input from a predetermined input device (not shown), the reading processing unit 7 reads the type characters, handwritten characters and marks written on the form 1. The form 1 is provided with a plurality of predetermined fields, and characters or the like are entered in each of the fields. The reading processing unit 7 reads characters and the like written in each field as the contents of each field. The reading processing unit 7 includes a plurality of forms 1
When the reading is completed, the result of the reading for each form 1 is stored in a predetermined storage area, and the character recognition processing unit 8 is activated.

Here, as the form 1, for example, the form shown in FIG. 6 is used. Therefore, this form 1 has a membership number, application date, postal code, telephone number, address (kana), address (kanji), name (kana), name (kanji), date of birth, occupation, year in order from the top. The age and gender are entered in predesignated fields.

As described above, only the member number is the print field and the others are the handwritten fields. In the handwritten field, application date, zip code, telephone number, date of birth,
Occupation and age fields are numeric fields, date of birth and gender fields are mark fields, address kana and name kana fields are kana fields, and address kanji and name kanji fields are kanji fields. Is. The application date, postal code, address, name, etc. are attributes of the data (that is, knowledge) entered in the respective fields.

When the character recognition processing unit 8 is activated by the reading processing unit 7, the character recognition processing unit 8 performs character recognition processing on the reading results of a plurality of sheets 1 stored in a predetermined storage area, and candidates for the character recognition results are obtained. Get the character string 9. The candidate character string 9 is stored in a predetermined file in a predetermined format. The candidate character string 9 is obtained for each form 1. The candidate character string 9 (one candidate character string 9) for one form 1 is composed of candidate character strings for each field of one form 1. When the character recognition processing unit 8 obtains the candidate character string 9 by character recognition of the form 1, the knowledge processing unit 11 is activated to perform the knowledge processing on the candidate character string 9. The knowledge processing unit 11 may be included in the character recognition processing unit 8.

When the knowledge processing unit 11 is activated by the character recognition processing unit 8, the knowledge processing unit 11 performs knowledge processing on the candidate character string 9 to obtain the character recognition output 6. This knowledge processing is performed for each one candidate character string 9 and is performed for each candidate character string 9 in each candidate field as a processing unit.

The knowledge processing section 11 is a kana-kanji automatic cooperation processing section 3A for performing preprocessing of knowledge processing, an independent knowledge processing section 4 and an associated knowledge processing section 5 for performing knowledge processing according to the result of the preprocessing, mainly an independent knowledge processing section. 4 and the associated knowledge processing unit 5 includes attribute data 10 and a dictionary 14 used for knowledge processing.

The kana-kanji automatic cooperation processing unit 3A corresponds to the automatic cooperation processing unit 3 of FIG. 1 and, as a preprocessing of the knowledge processing, the candidate character strings for each field are targeted for the knowledge processing or the independent knowledge processing. Or the target of the collaborative knowledge processing is determined.

First, the kana-kanji automatic cooperation processing unit 3A determines whether or not the candidate character string for each field of the candidate character string 9 is to be the target of knowledge processing. In this determination processing, the Kana-Kanji automatic cooperation processing unit 3A refers to the attribute data 10 which is originally data used by the independent knowledge processing unit 4 and the cooperation knowledge processing unit 5 for knowledge processing.

The attribute data 10 is prepared in advance for each type of form 1. That is, as shown by the dotted line in the figure, the type of the form 1 and the attribute data 10 have a one-to-one correspondence. Attribute data 1
0 stores the attribute for each field in one type of form 1. The attributes to be stored are the type of handwriting / printing, which is determined by the writing method, the type of kana / kanji / numeric / alphabetic / mixed, which is the attribute determined by the type of written characters, and the type 1 of the field There is necessity of position information, which is an attribute determined by the position, and knowledge processing, which is an attribute determined by whether knowledge processing is required for the field.

For example, in the form 1 shown in FIG. 6, the attribute of the member number field is "print type, number, position information, knowledge processing not required". Application date, zip code, phone number,
The attributes of the fields of birth (first half), occupation, and age are "handwriting, numbers, position information, and knowledge processing is not required". The attributes of the fields of birth (second half) and gender are "handwriting, marks, position information, and no knowledge processing required". The attributes of the fields of address kana and name kana are "handwriting, kana, position information, knowledge processing required". The attributes of the address Kanji and name Kanji fields are "Handwriting, Kanji,
Position information and knowledge processing are required ”.

In addition to the above, the item number of each field (for example, the item number of the member number field is "1") and the attribute (knowledge attribute) determined by the content of the data (knowledge) entered in each field. ) Is the "membership number"
Etc. are described as attributes of each field.

Such attribute data 10 is input / output device (not shown) included in the character recognition processing device in the recognition information definition process executed prior to the start of the character recognition process.
Is input in advance by the operator from and is stored in a predetermined file. For example, when the operator reads the blank form 1, the character recognition processing device displays it on the display screen of the input / output device. Therefore, the operator defines each field from this screen using a mouse or the like, and inputs each attribute (excluding position information) for each defined field from the input / output device. A predetermined processing program creates and manages such attribute data 10.

The position information in the attributes is automatically set by the processing program. That is, when the operator defines each field, the X coordinate (position) and the Y coordinate (position) of the point input on the screen are used as the position information of the field.

This position information will be described with reference to FIG. FIG. 3 shows an enlarged part of the form 1 shown in FIG. The position information on the form 1 is defined by the X coordinate (position) and the Y coordinate (position) with reference to the mark 1A. In the figure,
The horizontal direction is the X direction, and the vertical direction (search direction) is the Y direction. When the mark 1A is not printed, the end of the form 1 is set as a reference. That is, the position of each field on the form 1 is the start X position of the mark 1A.
The distance from _S, is represented by the distance from Y _S is the starting Y position of the mark 1A.

For example, in FIG. 3, the start X position of the address kana field, which is the fifth field (item number 5) from the beginning of the form 1, is X ₅₁ , and the start Y position is Y _51.
And the end Y position is Y ₅₂ . Note that the end X position X
₅₂ (not shown) is similarly determined. The same applies to other fields. The start X position and the start Y position of the position information indicate the position where the field is started when the form 1 is searched in the X direction and the Y direction, respectively.

When the form 1 as shown in FIG. 3 is displayed on the display screen of the input / output device, the operator operates the upper left corner 1B and the lower right corner (diagonal of the upper left corner 1B, not shown) of the field. Position the mouse on and specify the address kana field. At this time, the processing program sets the start X position, the start Y position, the end X position, and the end Y position of the address kana field as position information of the attribute data 10
Write in. This is done for each field. Thereby, the position information of each field is automatically set by the processing program.

With reference to such attribute data 10, the kana-kanji automatic linkage processing unit 3A knows, among the fields of the form 1, a field that is handwritten and has kana characters or kanji written therein. Target for processing. That is, in the handwritten kana field and the handwritten kanji field, it is difficult to increase the character recognition rate only by the character recognition process because the type of characters to be entered is large and the character shape depends on the personality of the writer. The candidate character string 9 is corrected. On the other hand, the kana-kanji automatic cooperation processing unit 3A excludes the type field and the number field from the knowledge processing targets.

For example, in the form 1 shown in FIG. 6, since the attribute of the member number field is "print", it is excluded from the knowledge processing target, and the application date, postal code, telephone number,
The birth date, occupation, age, and sex fields are also excluded because their attributes are "numbers". On the other hand, the address kana and name kana fields, which are handwritten kana fields, and the address kanji and name kanji fields, which are handwritten kanji fields, are subjected to knowledge processing.

Next, the kana-kanji automatic cooperation processing unit 3A determines whether the candidate character string for each field determined to be the target of the knowledge processing is to be the target of the independent knowledge processing or the cooperative knowledge processing. That is, it is determined whether or not the fields on the form 1 are to be linked with other fields. Fields that have not been subjected to cooperative knowledge processing are subject to independent knowledge processing.

The field to be the target of the collaborative knowledge processing is a field which is handwritten and in which kana or kanji is entered. For example, in the form 1 shown in FIG. 6, the address kana field and the address kanji field, and the name kana field and the name kanji field are targets to be linked.

In the form 1 shown in FIG. 6, all the fields to be the target of the knowledge processing are the fields to be the target of the collaborative knowledge processing, but depending on the type of the form 1, the target of the knowledge processing. In some cases, the fields that should be included include fields that should not be the target of the cooperative knowledge processing.

Here, as can be seen from FIGS. 6 and 3,
The fields to be subjected to the cooperative knowledge processing have predetermined positions on the form 1, and are located at extremely close positions or adjacent to each other. Therefore, the kana-kanji automatic cooperation processing unit 3A searches the plurality of fields on the form 1 for a kana field and a kanji field whose positions on the form 1 have a predetermined relationship. That is, the search is performed with "the positions on the form 1 having a predetermined relationship" as the search condition. Then, as a result of the search, the extracted kana field and the kanji field are linked. In particular, the handwritten kana field and the handwritten kanji field having such a positional relationship are linked.

In order to search a kana field and a kanji field whose positions on the form 1 have a predetermined relationship, the kana-kanji automatic cooperation processing section 3A recognizes the horizontal writing recognition field search processing section 12 and the vertical writing recognition. A field search processing unit 13 is provided.

Horizontal writing recognition field search processing unit 12
Searches for fields to be linked when the form 1 is horizontally written as shown in FIGS. 6 and 3. The horizontal writing recognition field search processing unit 12 searches for the horizontal writing form 1 from the upper side to the lower side (Y direction), and finds the start X of each other on the form 1 from among the plurality of fields on the form 1. A kana field and a kanji field whose positions are substantially the same and whose starting Y position of the lower field is extremely close to the upper field are searched. That is, the search is performed under the search condition that "the starting X positions of the forms 1 are substantially the same on the form 1 and the starting Y positions of the lower fields are extremely close to the upper fields".

For example, in the form 1 of FIG. 3, the address Kana field and the address Kanji field are the start X of each other.
The positions X ₅₁ and X ₆₁ are the same, and the starting Y position Y ₆₁ of the lower address Kanji field is extremely close to the ending Y position Y ₅₂ of the upper address Kana field. The same applies to the name kana field and the name kanji field. Therefore, these are searched as fields to be linked.

Whether or not the start X positions are substantially the same is determined by the start X of the address Kana field and the address Kanji field.
This is performed depending on whether or not the position difference (X ₆₁ -X ₅₁ ) is smaller than the predetermined value A. The predetermined value A is an extremely small value,
Given in advance.

Further, the judgment as to whether or not the start Y position is extremely close is made by determining the difference (Y ₆₁ -Y ₅₂ ) between the end Y position of the address kana field and the start Y position of the address kanji field as a predetermined value. It is performed depending on whether it is smaller than B or not. Predetermined value B
Is given in advance. The predetermined value B is, for example, a value (Y ₆₂ -Y ₆₁ ) or (X ₅₂ -X ₅₁ ) on one side of the writing area for one character. Note that the predetermined value A is a value that is sufficiently smaller than the predetermined value B.

The conditions for searching the fields to be linked are not limited to the above search conditions. For example, the handwriting kana field and the handwriting kanji field may have the same knowledge attribute as the search condition. Further, on the form 1, the handwritten kana field and the handwritten kanji field may be adjacent to each other as a search condition.
Further, these search conditions may be changed as needed.

Horizontal writing recognition field search processing unit 12
Performs this search as follows: In order to search the form 1 in the Y direction, the attribute data 10 is searched using the Y position of the search target.
Refer to. When the start Y position of each field is equal to the search target Y position, the field is searched. The search target Y position is incremented from the value Y _S. Note that, for example, when there are a plurality of fields having the same starting Y position, such as the membership number field and the application date field, the starting X position of the attribute data 10 is further referred to, and a field having a smaller value (for example, the membership number field). ) Is searched first.

Vertical writing recognition field search processing unit 13
Searches for fields to be linked when the form 1 is vertically written. The vertical writing recognition field search processing unit 13 searches for a vertical writing form from the right to the left, and selects a form 1 from a plurality of fields on the form.
The starting Y positions of each other are substantially the same, and
The Kana field and the Kanji field in which the start X position of the field on the left side is extremely close to the field on the right side are searched. In other words, the search is performed with the search condition that "the start Y positions of the two are substantially the same on the form 1 and the start X position of the left field is extremely close to the right field". To do.

For example, consider a form in which the form 1 in FIG. 3 or FIG. 6 is rotated 90 degrees so that the member number field becomes a vertical writing area at the right end of the form 1. X direction and Y
The direction is opposite to that in the example of FIG. 3 or FIG. In such a form, the address Kana field and the address Kanji field have the same starting Y position and the starting X position.
The position is extremely close. The same applies to the name kana field and the name kanji field. Therefore, these are searched as fields to be linked. The search condition and the search method are the same as in the above case.

Through the above processing, the kana-kanji automatic linkage processing section 3A links the address kana field and the address kanji field among the fields subjected to the knowledge processing, and creates the name kana field and the name kanji field. Make them work together.

The Kana-Kanji automatic cooperation processing unit 3A performs this cooperation processing as follows. For convenience, the previously searched address kana field is set as the side to be linked, and the subsequently searched address kanji field is set as the side to be linked. Attribute data 10
In step 3, the cooperation flag is set in the attribute of the address kana field of the cooperation side, and the item number “6” of the address kanji field of the cooperation side is set. The same applies to the name field.

The kana-kanji automatic cooperation processing unit 3A transfers control to the independent knowledge processing unit 4 or the cooperation knowledge processing unit 5 according to the result of the above preprocessing. The independent knowledge processing unit 4 or the cooperative knowledge processing unit 5 uses the dictionary 14 to perform knowledge processing.

The dictionary 14 includes a name kana dictionary, a name kanji dictionary, an address kana dictionary, and an address kanji dictionary. For example, the name / kana dictionary is a dictionary for knowledge processing of the name / kana field, and is composed of data in which existing names and names are described in kana. The same applies to the name / kanji dictionary, address kana dictionary, and address kanji dictionary.

The independent knowledge processing unit 4 performs the knowledge processing in the character recognition processing for a predetermined field by the field alone. That is, in the Kana-Kanji automatic cooperation processing unit 3A, the following independent knowledge processing is performed for fields that are not cooperated with other fields.

When the independent knowledge processing unit 4 is given control from the Kana-Kanji automatic cooperation processing unit 3A together with the item number of the field to be processed, the independent knowledge processing unit 4 carries out independent knowledge processing for the field indicated by the item number.

That is, the candidate character string obtained from the field is extracted from the candidate character string 9 using the item number. Also,
By referring to the attribute of the item number of the attribute data 10, the attribute is known. Therefore, the extracted candidate character string is used to collate the corresponding dictionary 14 to check whether or not the candidate character string exists. If the candidate character string exists, the candidate character string is left as it is. Correction is performed using the columns as candidate character strings.

The cooperative knowledge processing unit 5 performs knowledge processing in character recognition processing for a predetermined field by coordinating that field with other fields associated with this field. That is, the following collaborative knowledge process is performed on the handwritten kana field and the handwritten kanji field that are collaborated by the kana-kanji automatic collaboration processing unit 3A.

Specifically, the kana-kanji automatic cooperation processing unit 3A
When the control is passed from the item together with the item number of the field to be processed, cooperative knowledge processing for the field indicated by the item number is performed.

For example, when the item number is “7” indicating the name and kana field, the candidate character string obtained from the name and kana field is extracted from the candidate character string 9. Further, the attribute (name and kana field attribute) of the item number “7” of the attribute data 10 is referred to, and the name and kana are known as the attributes.
Further, it is known that the cooperation flag is set as the attribute of the name and kana field and that the object of cooperation is the name and kanji field of the item number “8”.

Therefore, the extracted candidate character string is used to collate the name kana dictionary to check whether the candidate character string exists. If it exists, leave it as it is. If it does not exist, it exists in the name kana dictionary. Corrects a character string that is close to the candidate character string (performs independent knowledge processing). For example, assume that the result is "Tanaka".

Next, the candidate character string obtained from the name / kanji field to be linked is extracted from the candidate character string 9 and the attribute is known from the attribute data 10. Then, using the extracted candidate character string, the name / kanji dictionary is checked to see if the candidate character string exists. If it exists, leave it as it is. If it does not exist, close characters that exist in the name / kanji dictionary. Correction is performed using strings as candidate character strings (single knowledge processing is performed). For example, assume that this result is "Yunaka".

Here, it is assumed that the candidate character string "Tanaka" obtained from the name kana field is correct and the candidate character string "Yanaka" obtained from the name kanji field is incorrect. Further, it is assumed that these character strings are both present in the name kana dictionary and the name kanji dictionary. In this case, the candidate character string cannot be corrected by the independent knowledge processing.

In the cooperation knowledge processing unit 5, the candidate character string "Tanaka" obtained from the name Kana field has three characters, while
It is determined that the candidate character string "Tanaka" is correct based on the fact that the reading of the candidate character string "Yunaka" obtained from the name and kanji field becomes four characters "Yoshinaka". This is because normally, the number of characters of the candidate character string obtained from the name kana field cannot be wrong.

Next, the cooperation knowledge processing unit 5 refers to the name / kanji dictionary by using the candidate character string “Tanaka” which is determined to be correct, and creates a character shape close to the candidate character string “Yanaka” obtained from the name / kanji field. The character string "Tanaka" whose reading is "Tanaka" is corrected to be a candidate character string.

The above description of the knowledge processing is as follows.
This is an example, and in practice, the independent knowledge processing unit 4 and the cooperative knowledge processing unit 5 perform other various kinds of knowledge processing. FIG. 4 is a knowledge processing flow, and shows processing performed by the kana-kanji automatic cooperation processing unit 3A and the horizontal writing recognition field search processing unit 12. In this processing flow, when searching the form 1, the handwritten kana field is searched before the handwritten kanji field among the fields to be linked,
In addition, a case where the form 1 in which these are continuously searched is processed will be described.

When the character recognition processing unit 8 starts the knowledge processing unit 11, the knowledge processing unit 11 processes the form 1 shown in FIG.
It is checked whether or not it is the horizontal writing as shown in (S1). If horizontal writing is not possible, the processing shown in FIG. 5 is performed. In the case of horizontal writing, control is passed to the horizontal writing recognition field search processing unit 12.

The control field for horizontal writing recognition field 12 to which the control is passed searches the form 1 to be processed from the top to the bottom for the presence or absence of a field (S2).
When a field is detected by this search, the horizontal writing recognition field search processing unit 12 refers to the attribute data 10 to check whether or not the field is a handwritten Kana field (S3).

If it is not a handwritten Kana field, the horizontal recognition field search processing unit 12 notifies the knowledge processing unit 11 of this fact, and the knowledge processing unit 11 checks whether the field searched in S2 is the final field (S4). ). Therefore, the maximum value of the item number of the form 1 is set in a predetermined register in advance, and 1 is subtracted from the contents of the register every time the field is searched in S2 and S6. Register contents are 0
If it becomes, it is regarded as the final field.

If it is the final field, the knowledge processing unit 1
1 ends knowledge processing. If not the last field,
The knowledge processing unit 11 is a horizontal recognition field search processing unit 1.
Control is passed to step 2, and steps S2 and below are repeated.

In the case of a handwritten Kana field,
The recognition field search processing unit 12 for horizontal writing refers to the attribute data 10 to check whether the field searched in S2 is a knowledge field (S5).

If it is not a knowledge field, S4 and subsequent steps are executed for the field retrieved in S2. If the field is a knowledge field, the horizontal writing recognition field search processing unit 12 searches the form 1 further downward from the position of the field searched in S2, and searches for the next field (S6). The next field is usually the X of the field whose start X position was retrieved in S2, as shown in FIG.
The position is almost the same as the position, and the starting Y position is extremely close to the Y position of the field searched in S2.

When the next field is detected by this search, the horizontal writing recognition field search processing unit 12 refers to the attribute data 10 to check whether or not the next field is a handwritten Chinese character field (S7).

If it is not a handwritten Chinese character field, the horizontal recognition field search processing unit 12 transfers control to the independent knowledge processing unit 4, and the independent knowledge processing unit 4 determines that the field searched in S2 and the field searched in S6 are: Each independently executes independent knowledge processing (S8). After S8 ends, S4 and subsequent steps are executed for the field retrieved in S6, which is the later retrieved field.

In the case of a handwritten Kanji field, further
The horizontal writing recognition field search processing unit 12 refers to the attribute data 10 to check whether the next field is the knowledge field and the knowledge attribute is the same as the field searched in S2 (S9).

If they are knowledge fields and the knowledge attributes are not the same, S8 is executed. That is, the horizontal writing recognition field search processing unit 12 transfers control to the independent knowledge processing unit 4, and the independent knowledge processing unit 4 independently performs independent knowledge processing for each of the field searched in S2 and the field searched in S6. Is executed (S8). After the end of S8,
S4 and the following steps are executed for the field retrieved in S6, which is the field retrieved later.

If the knowledge fields are the same and the knowledge attributes are the same, the horizontal recognition field search processing unit 12
Notifies the effect to the Kana-Kanji automatic cooperation processing unit 3A, and the Kana-Kanji automatic cooperation processing unit 3A cooperates the Kanji field searched in S6 with the Kana field searched in S2 (S10).

Then, the horizontal writing recognition field search processing unit 12 transfers control to the cooperation knowledge processing unit 5, and the cooperation knowledge processing unit 5 executes cooperation knowledge processing (S11). After this cooperative knowledge processing, the field S6 which is the one retrieved later is used.
The step S4 and the following steps are executed for the field retrieved in step S4.

FIG. 5 is a knowledge processing flow. The kana-kanji automatic cooperation processing section and the vertical writing recognition field search processing section 1
3 shows the processing performed by 3. This processing flow shows processing that is executed subsequent to S1 in FIG. Therefore, the case where the handwritten kana field is searched before the handwritten kanji field and these are continuously searched is processed.

Form 1 to be processed in S1 of FIG.
When it is recognized that is written vertically, the Kana-Kanji automatic cooperation processing unit 3A transfers control to the vertical writing recognition field search processing unit 13.

The vertical writing recognition field search processing unit 13 to which the control is passed searches the form 1 to be processed from the right to the left for the presence or absence of a field (S1).
2). When the field is detected by this search, the vertical writing recognition field search processing unit 13 checks whether or not the field is a handwritten Kana field, as in S3 (S13).

If it is not the handwritten Kana field, the knowledge processing unit 11 checks whether or not the field searched in S12 is the final field, as in S4 (S14). If it is the final field, the knowledge processing unit 11 ends the knowledge processing. If it is not the final field, the knowledge processing unit 11 passes control to the vertical writing recognition field search processing unit 13,
The steps from S12 onward are repeated.

In the case of a handwritten Kana field, further
Similar to S5, the vertical writing recognition field search processing unit 13 checks whether or not the field searched in S12 is a knowledge field (S15).

If it is not the knowledge field, the steps from S14 are executed. If it is a knowledge field, the vertical writing recognition field search processing unit 13 turns the form 1 further leftward from the position of the field searched in S12, and searches for the presence or absence of the next field (S16). The next field is
The starting Y position is substantially the same as the Y position of the field searched in S12, and the starting X position is extremely close to the X position of the field searched in S12.

When the next field is detected by this search, the vertical writing recognition field search processing unit 13 checks whether or not the next field is a handwritten Chinese character field as in S7 (S17).

If it is not a handwritten Chinese character field, the vertical writing recognition field search processing unit 13 transfers control to the independent knowledge processing unit 4, and the independent knowledge processing unit 4 executes the independent knowledge processing (S18). After that, S14 and subsequent steps are executed.

In the case of a handwritten Chinese character field,
As in S9, the vertical writing recognition field search processing unit 13 checks whether or not the next field is the knowledge field and the knowledge attribute is the same as the field searched in S12 (S19).

If they are knowledge fields and the knowledge attributes are not the same, S18 is executed. That is, the independent knowledge processing unit 4 executes the independent knowledge processing (S18). If the knowledge fields are the same and the knowledge attributes are the same, the Kana-Kanji automatic linkage processing unit 3A links the Kanji field searched in S16 and the Kana field searched in S12, as in S10 (S20). .

After that, the vertical writing recognition field search processing unit 13 activates the cooperation knowledge processing unit 5, and the cooperation knowledge processing unit 5
Executes cooperative knowledge processing (S21). After this, S12
Do the following:

The processing flows shown in FIGS. 4 and 5 show an example of automatic cooperation processing in the knowledge processing.
Various changes can be made according to each attribute of the fields in the form 1. That is, this processing flow is for processing the form 1 in which the handwritten kana field is searched before the handwritten kanji field and these are continuously searched. Therefore, in the form 1, the processing flow is changed when the order of detection is reversed or when continuous search is not performed.

Next, the cooperation process of the forms shown in FIGS. 3 and 6 according to the process flow of FIG. 4 will be briefly described. Since this form 1 is written horizontally, the knowledge processing unit 1
1 recognizes that the form 1 to be processed is horizontal writing in S1 and passes control to the horizontal writing recognition field search processing unit 12. First, the recognition field search processing unit 12 for horizontal writing detects the member number field in S2, but since this is a type field and not a handwritten Kana field (S3), no knowledge processing is performed. next,
After S4, the application date field is detected in S2, but since this is also a numeric field and not a handwritten Kana field (S3), knowledge processing is not performed. The same applies to the postal code field and the telephone number field.

Next, the horizontal writing recognition field search processing section 12 detects the address kana field in S2.
Since this is a handwritten kana field (S3) and a knowledge field (S5), another field is searched downward in S6 to detect the address Kanji field. This is a handwritten Kanji field (S
7) and the knowledge attribute is the same for "address" (S
9), the recognition field search processing unit 12 for horizontal writing, S1
In 0, the address Kanji field and the address Kana field are linked. As a result, cooperative knowledge processing is performed for the address Kanji field and the address Kana field (S11).

Next, the horizontal writing recognition field search processing section 12 detects the name kana field in S2.
Since this is a handwritten kana field (S3) and a knowledge field (S5), another field is searched downward in S6 to detect the name / kanji field. This is a handwritten Kanji field (S
7) and since the knowledge attribute is the same for "name" (S)
9), the recognition field search processing unit 12 for horizontal writing, S1
In 0, the name kanji field and the name kana field are linked. As a result, cooperative knowledge processing is performed for the name / kanji field and the name / kana field (S11).

[0094]

As described above, according to the present invention,
In the character recognition processing device, the handwritten kana field and the handwritten kanji field are automatically calculated by utilizing the fact that the positions of the handwritten kana field and the handwritten kanji field, which should be subject to cooperative knowledge processing, have a predetermined relationship on the form. Since it is possible to perform automatic search, it is possible to automatically link these in most forms and perform cooperative knowledge processing. This reduces the burden on the operator without the need to input instructions for cooperative knowledge processing. It is possible to prevent the occurrence of an error in the instruction input, and improve the efficiency of character recognition processing.

[Brief description of drawings]

FIG. 1 is a principle configuration diagram of the present invention.

FIG. 2 is a configuration diagram of an embodiment.

FIG. 3 is an explanatory view of an embodiment.

FIG. 4 is a knowledge processing flow.

FIG. 5 is a knowledge processing flow.

FIG. 6 is a background explanatory diagram of the present invention.

[Explanation of symbols]

 1 form 2 processing device 3 automatic cooperation processing unit 3A kana-Kanji automatic cooperation processing unit 4 independent knowledge processing unit 5 cooperation knowledge processing unit 6 character recognition output 7 reading processing unit 8 character recognition processing unit 9 candidate character string 10 attribute data 11 knowledge processing Part 12 Horizontal writing recognition field search processing unit 13 Vertical writing recognition field search processing unit 14 Dictionary

Claims (3)

[Claims] 1. A character written in each field of a form having a plurality of predetermined fields is read,
In a character recognition processing device that performs character recognition processing for this, a knowledge processing in character recognition processing for a predetermined field is performed by an independent knowledge processing unit that performs the field alone and a knowledge processing in character recognition processing for a predetermined field. , A field for collaborative knowledge processing that links the field with other fields linked to this field, and searches for a plurality of fields whose positions on the form have a predetermined relationship from among the plurality of fields. However, a character that is provided with an automatic cooperation processing unit for coordinating these, and for the plurality of fields that are cooperated by the automatic cooperation processing unit, the cooperation knowledge processing unit performs the knowledge processing by coordinating them Recognition processing device. 2. The character according to claim 1, wherein the automatic cooperation processing unit includes a Kana-Kanji automatic cooperation processing unit, and the Kana-Kanji automatic cooperation processing unit cooperates a handwritten Kana field and a handwritten Kanji field. Recognition processing device. 3. The kana-kanji automatic linkage processing unit searches for a form for horizontal writing from top to bottom,
From a plurality of fields on the form, a handwritten Kana field whose start X positions are substantially the same on the form and the start Y position of the lower field is extremely close to the upper field. The recognition field search processing unit for horizontal writing that searches the handwritten Kanji field, and the form for vertical writing are searched from the right to the left,
From the plurality of fields on the form, the start Y positions of the two are substantially the same on the form, and the start X position of the left field is very close to the right field. The character recognition processing device according to claim 2, further comprising: a vertical text recognition field search processing unit that searches the field and the handwritten Chinese character field.