JPH02311974A - Character recognizing device - Google Patents

Abstract

PURPOSE:To discriminate characters necessary or unnecessary for processing and to prevent following processing from being affected by the unnecessary character by providing a recognizing means, unnecessary character storing means, comparing means to compare the respective characters, which constitute a character string, with the unnecessary character and character string constituting means. CONSTITUTION:The character unnecessary for the processing is selected in advance and stored in an unnecessary character storing means 57. The respective characters constituting the character string, which is recognized by a recognizing means 53, are compared whether or not it correspond to the unnecessary character, which is stored in the unnecessary character storing means 57, by a comparing means 54. As the result of this comparison, the character corresponding to the unnecessary character is deleted from the character string by a character string constituting means 55 and the new character string is constituted. Accordingly, since the unnecessary character can be recognized and deleted from the character string without fail, only the necessary character can be picked up securely without being affected by the arrangement of the characters. Thus, the character string composed only of the necessary character string can be obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a character recognition device that deletes unnecessary characters from a character string.

(Prior Art) When conducting various transaction operations (exchange operations, etc.) in financial institutions, etc., predetermined forms are used. After the customer fills in the necessary information on the form, the customer requests a bank employee to process the form.The bank employee then performs a predetermined process according to the contents entered on the form.

For example, in the case of money transfer processing, a money order form as shown in FIG. 2 is used.

Figure 2 shows a surface diagram of the exchange form.

The money order form 1 shown in the figure has three items: a payee item 2, a recipient item 3, and a client item 4.

The transfer destination item 2 is composed of a financial institution type entry field 2a, a bank name entry field 2b, and a branch name entry field 2C. Beneficiary item 3 is account number entry field 3a and name entry field 3b
and an amount entry field 3C. The client item 4 is composed of a name entry field 4a and a client code entry field 4b.

The financial/institution type entry column 2a of the transfer destination item 2 is a column where you select the desired option from among the options and mark it with a circle, and the other columns consist of multiple squares in which you can write characters. has been done.

When performing money transfer operations using the above-mentioned money order form l, a money order data communication device is used.

FIG. 3 shows a block diagram of a conventional exchange data communication device.

The exchange data communication device 10 shown in the figure includes a sales branch 11a,
It is comprised of facsimile machines 12a and 12b installed in the ILB, a facsimile 14 installed in the district center 13, terminal equipment 15a and 15b, and a host device 16.

Sales branch 11a, llb facsimile (FAX) 12a
, 12b are connected to a facsimile (EAX) 14 at the district center 13 via a public communication network 17. Terminal device 15a of district center 13.

1'5b is connected to the host device 1 via the dedicated communication line 18.
6. In addition, the business store 11a.

Reference numeral 11b denotes a store where bank employees are placed to process various transaction requests from customers. The district center 13 is an information collection point that compiles information from a plurality of branches. The host device 16 is composed of a host computer and the like.

Now, in the exchange data communication device IO with the above configuration,
Foreign exchange operations are carried out using the following steps.

■The sales branch 11a sends a money order form with the specified information filled in.
(shown in FIG. 2) by the facsimile 12a, and the branch Ilb transmits the exchange form I by the facsimile 12b to the facsimile 14 of the district center 13 via the public communication network 17.

- At the district center 13, the facsimile exchange forms (FAX exchange forms) la (exchange forms made of facsimile paper) outputted from the facsimile machine 14 are manually sorted into branch units.

(2) The sorted facsimile money order forms 1a are input manually using the terminal devices 15a and 15b, and are transmitted as money order data to the host device 16 via the dedicated communication line 18.

■The host device 16 performs processing based on the received exchange data and completes the work.In the case of the above-mentioned exchange data communication device 10, a worker and a terminal sort the facsimile exchange forms 1a in the district center 13. Equipment 15a.

A problem has arisen in that a large number of workers must be allocated to operate the 15b. In addition, the terminal devices 15a, 1
5b, there is a possibility that an input error may occur, and furthermore, there are problems in that a considerable amount of time must be spent on data input.

Therefore, a money exchange data communication device using a character recognition device has been proposed.

FIG. 4 shows a block diagram of a money exchange data communication device using a conventional character recognition device.

In the exchange data communication device 20 shown in the figure, the branch 2
1 is an optical character reader (OCR) 22 and this 0
It consists of a keyboard 23 and a network control unit (NCU) 24 connected to the CR 22. The district center 26 includes a character recognition device 27,
A network control unit (NCU) 28 connected to this character recognition device 27 and a terminal device 29a
, 29b, the zero character recognition device 27 is provided with a storage device 27a. NC of branch 21
The U 24 connects to the N of the district center 26 via the public communication network 30.
It is connected to CU28. The character recognition device 27 of the district center 26 is connected to a host device 32 via a dedicated communication line 31.

The OCR 22 of the branch 21 optically reads the characters written on the exchange form 1 and outputs the image data thereof. The keyboard 23 is used to input kana and numerical values. The NCU 24 is composed of a modem and the like that converts data within the branch 21 into an audible signal for transmission to the public communication network 30.

The character recognition device 27 of the district center 26 converts the data (image data) of the exchange form 1 received from the branch 21 into a character string (recognition result) indicated by the character code, and further converts the recognition result to the host device. This process converts the data into data for transmission to be sent to 32. The storage device 27a is composed of a hard disk device or the like that stores a conversion table that is referred to when the character recognition device 27 converts recognition results into data for transmission. This conversion table associates the recognition result with the data for transmission. For example, if the recognition result is a kanji character and the data for transmission is a kana character, the conversion table is composed of a table that associates the kanji character with the kana character.

Like the NCU 24, the NCU 28 is composed of a modem and the like. The terminal devices 29a and 29b are used to confirm the recognition result recognized by the character recognition device 27 and transmission data to be transmitted to the host device 32, and to instruct the host device 32 to transmit the recognition result.

Now, in the money transfer data communication device 20 having the above configuration, money transfer operations are performed according to the following procedure.

■The branch 21 sends the exchange form l (shown in Figure 2) as 0CR.
22 to read it.

(2) The NCU 24 of the branch 21 transmits image data based on the content read by the OCR 22 to the NCU 28 of the district center 26 via the public communication network 30.

(2) The NCU 28 of the district center 26 transfers the received image data to the character recognition device 27.

(2) The character recognition device 27 recognizes the pattern of the transferred image data and converts it into a character string (recognition result) indicated by a character code.

■The character recognition device 27 converts the recognition result into transmission data to be sent to the host device 32 by referring to the conversion table provided in the storage device 27a, and uses this transmission data and the image received from the branch 21 For example, the data is transferred to the terminal device 2.9a.

(2) The terminal device 29a displays the transferred image data and transmission data on its display screen.

(2) After confirming the displayed image data and transmission data, the operator instructs the host device 32 to transmit the transmission data. This transmission data is transmitted from the character recognition device 27 to the host device 32 via the dedicated communication line 31. Note that the transmission data sent to the host device 32 is composed of kana characters and character codes of numerical information.

(2) The host device 32 executes a predetermined exchange process based on the received data and ends the transaction.

The exchange process was carried out using the procedure described above.

(Problem to be solved by the invention) Here, the operation of the character recognition device 27 will be explained in detail.

FIG. 5 shows an explanatory diagram of the operation of the conventional character recognition device 27.

The figure shows three correspondences: characters written on the exchange form 1 (FIG. 2), recognition results by the character recognition device 27, and transmission data converted by the character recognition device 27.

Here, the branch name entry field 2c and amount entry field 3C of the money order form 1 will be explained as examples.

It is also assumed that the branch name entry column 2C is made up of cells corresponding to 5 characters, and the amount entry column 3c is made up of cells made up of 10 characters. Furthermore, in the storage device 27a, “Tokyo b
It is assumed that a conversion table is prepared that associates "bb" with "Tokyo". Note that "b" indicates a blank.

Now, item number l is "Tokyo" in the branch name entry column 2c, aligned to the left.
This is the case where the following is entered. In this case, the character recognition device 27 obtains a character string (recognition result) consisting of a character code for a total of five characters, two characters of “Tokyo” and three blanks (b). Specifically, the character recognition device 2 refers to the conversion table stored in the storage device 27a and converts the recognition result into transmission data.
7 accesses the storage device 27a and stores “Tokyo bbb” and “
The conversion table that associates "Tokyo bbb" with "Tokyo bbb" is thereby converted into transmission data of the pseudonym "Tokyo".

Next, as shown in Item No. 2, if you enter "Tokyo" in the branch name entry column 2c right-justified, the character recognition device 27 will not be able to read "bbb Tokyo" even if it refers to the conversion table in the storage device 27a. ” cannot be converted into the transmission data “TOKYO°°.” Therefore, the character recognition device 27 determines that there is a data error in the recognition result, and cancels the processing related to the exchange form l in which this data error occurred. This is because the conversion table in the storage device 27a does not have a table that associates "bbb Tokyo" with "Tokyo".Similarly, as shown in item number 3, the branch name entry field “East” in the 2nd and 4th squares of 2c
, “Kyo” is entered. In this case, the first, third, and fifth squares will be "blank (b)" because the conversion table for "bTokyob" is not provided in the storage device 27a. Character recognition device 27
Therefore, it is not possible to obtain data for transmission.

Therefore, a data error occurs in the character recognition device 27,
You will not be able to carry out any subsequent exchange processing.

Next, as shown in item number 4, "blank (b)" is written in the first and second squares of the amount entry column 3C, which is composed of squares for 10 characters, and from the third to the first O square. , “¥12
．． 345-" has been entered. In this case, the character recognition device 27 has two "blanks (b)" and "¥1
9 character recognition device 27 that obtains a character string (recognition result) consisting of character codes for a total of 10 characters, 8 characters of 2.345-”
If the recognition result is a character other than kanji, the recognition result is used as is as data for transmission. However, in the host device 32, non-numeric characters and
That is, since "blank b", "\", ", ", and "-" are included, a data error occurs.

Thus, in the case shown in item number 2.3.4, a data error occurs in the character recognition device 27 and the host device 32, and the exchange form 1 in which these data errors occur becomes invalid.

For this reason, in the past, it was necessary to prepare a large number of conversion tables in the storage device 27a, assuming various ways of using the squares when the customer first writes characters on the eyes. For this reason, a large-capacity storage device must be prepared, leading to problems such as increased costs. In addition, the characters entered in the amount entry field 3c are customarily "¥", "
, ", "-", etc. are often added. In such cases, a data error occurs only after the data is sent to the host device 32, so the data between the district center 26 and the host device 32 is There was a problem that communication was wasted.In such cases, it was necessary to create a new exchange form to match the specified entry method.For this reason, in the past, customers had to Please fill in the specified character entry method, for example, left-justified, and for numbers, enter “￥
This required troublesome restrictions and effort, such as not allowing characters such as “,”, etc. to be added.

The present invention has been made with attention to the above points, and it is possible to distinguish between characters unnecessary for processing and characters necessary for processing during character recognition, and to recognize characters in which the influence of unnecessary characters does not occur in subsequent processing. The purpose is to provide a processing device.

(Means for Solving the Problems) The character recognition device of the present invention includes a recognition means for recognizing a character string consisting of two or more characters, and an unnecessary character storage means for storing unnecessary characters that do not need to be finally recognized. , a comparison means for comparing each character constituting the character string with the unnecessary character, and a character string organization means for deleting the unnecessary character from the character string based on the comparison result of the comparison means to obtain a new character string. It is equipped with the following.

(Operation) The above device selects in advance unnecessary characters that are unnecessary for processing and stores them in the unnecessary character storage means. Then, the comparison means compares each character constituting the character string recognized by the recognition means to see if it corresponds to an unnecessary character stored in the unnecessary character storage means. As a result of this comparison, characters corresponding to unnecessary characters are deleted from the character string by the character string organization means, and a new character string is created. As a result, unnecessary characters are deleted from the character string recognized by the recognition means, and a character string composed only of necessary character strings can be obtained.

(Embodiment) FIG. 1 is a block diagram showing an embodiment of a character recognition device of the present invention.

In the figure, image data from an OCR 22, etc. is input to the input side of a character recognition device 50, which is composed of a processor 51, a hard disk device, etc. Further, from the output side, output data consisting of character codes, image data, etc. is output.

The processor 51 includes a recognition means 53, a comparison means 54,
A character string organizing means 55 is provided. Storage device 52
It is comprised of a character pattern storage means 56, an unnecessary character storage means 57, and a conversion table storage means 58.

The recognition means 53 of the processor 5l recognizes the image data output from the OCR 22, converts it into a corresponding character code based on the data stored in the character pattern storage means 56, and creates a character string (multiple characters) corresponding to the image data. (array of character codes). The comparison means 54 compares each character constituting the character string converted by the recognition means 53 with the unnecessary characters stored in the unnecessary character storage means 57, and outputs a comparison result of match or mismatch. The character string compiling means 55 deletes characters for which the comparing means 54 has produced a matching comparison result from the character string recognized by the recognizing means 52 to compile a new character string. Note that the character string organizing means 55 also performs a process of converting the organized character strings into transmission data to be sent to the host device 32.

The character pattern storage means 56 of the storage device 52 stores character codes such as kana, numbers, and even kanji in association with their image data. Note that image data that is not stored in the character pattern storage means 56 is not recognized by the recognition means 53 and is output as is. The unnecessary character storage means 57 is
It stores unnecessary characters among the characters recognized by the recognition means 53, such as symbols such as blank (b) and hyphen (-), and is divided into first to third areas 57a to 57c. There is. In this embodiment, "blank (b)" and "
¥” is stored in the second area 57b, “blank (b)” and “,” are stored in the third area 57c, and “blank (b)” and “hyphen (1)” are stored in the third area 57c. It is assumed that the conversion table storage means 58 is the character string organization means 5.
It stores a conversion table that associates character strings organized by 5 (m-formation results) with data for transmission.For example, if the recognition result is kanji and the data for transmission is kana, this kanji and kana are stored. It consists of a table that maps .

FIG. 6 shows an explanatory diagram of the operation of the character recognition device 50 of the present invention.

The figure shows four characters: characters entered in the exchange form l (Figure 2), recognition results by the character recognition device 50, compilation results by the character recognition device 50, and transmission data converted by the character recognition device 50. Correspondence is shown.

Here, branch name entry column 2c of money order form l (Figure 2)
This will be explained using the amount entry column 3C as an example. In addition, branch name entry field 2C starts from the 5th character square, and amount entry field 30 starts from 1st.
It is assumed that it is composed of 0 character squares.

Now, as shown in item number 1, it is assumed that "Tokyo" is entered in the 5-character box of branch name entry field 2c, aligned to the right. In this case, the first to third items are “blank (b)”.
“Howler. Now, the recognition means 53 refers to the character pattern storage means 56 and converts these five characters into a character code,
Create a 5-character string (recognition result). Thereafter, the comparison means 54 compares the first area 57a of the unnecessary character storage means 57.
Refer to , and compare (forward) from the first character to the first valid character (the character for "east"). As a result, the first to third characters "blank (b)" match unnecessary characters. Next, tracing back from the fifth character to the first valid character (the character "Kyo") is compared with reference to the second area 57b of the unnecessary character storage means 57. As a result, since the fifth character is "Kyo", no actual comparison is performed. Finally, the space (middle) between the first valid character (the character "East °") and the last valid character (the character "Kyopa") is stored in the third area 57c of the unnecessary character storage means 57.
See and compare. As a result, since no characters exist between "Higashi" and "Kyo", no actual comparison is performed in this case either. Now, the characters that are determined by the comparing means 54 to match unnecessary characters are deleted by the character string organizing means 55. Therefore, a new character string consisting of the two characters "Tokyo" is compiled (composition result) 0 The character string composition means 55 refers to the conversion table storage means 58 and obtains data for transmission of the kana "Tokyo". .

Next, as shown in item number 2, it is assumed that "Higashi" and "Kyoto" are written in the second and fourth positions of the five-character square, respectively. In this case, the first, third, and fifth characters are blank (b). Now, the recognition means 53 refers to the character pattern storage means 56 and converts these five characters into a character code. to create a 5-character string (
recognition results). Thereafter, the comparison means 54 recognizes this character string and compares the first character to the fifth character three times with the unnecessary characters stored in the unnecessary character storage means 57. First, the first comparison is made in the first area 5 of the unnecessary character storage means 57.
7a and the first valid character from the first character (“east”
Compare characters up to As a result, the first character blank (b) matches the unnecessary character. Next, the second area 57b of the unnecessary character storage means 57 is referred to and compared starting from the fifth character up to the first valid character (the character "Kyo"). As a result, the fifth character blank (b) matches an unnecessary character. And finally, unnecessary character storage means 57
With reference to the third area 57c, the first valid character (the character "East") to the last valid character (the character "Kyoto") are compared. As a result, the third character blank (b) matches the unnecessary character. Characters that are determined to match unnecessary characters by the comparing means 54 are deleted by the single character forming means 55, and a new character string consisting of the two characters "Tokyo" is formed (m formation result). Based on the (m-formation result), data for transmission such as "Tokyo" is obtained.

Next, as shown in item number 3, "Blank (b)" is written in the first and second boxes of 10 characters in the amount entry field 3c, and "Blank (b)" is written in the 3rd to 10th boxes in order. ¥12.34
5-" is entered.

The recognition means 53 refers to the character pattern storage means 56 and creates a character string of 10 characters (recognition result), and then the comparison means 54 compares the first character string of the unnecessary character storage means 57.
This is done with reference to area 57a. In this case, first, the blank (
b)” and “¥” are determined to match.Next, the second area 57b of the unnecessary character storage means 57 is referred to, and the valid character “5” is found.
” is determined to be a match.Finally, the third area 57c of the unnecessary character storage means 57 is referred to, and the character located between the valid character “2” and the valid character “3” is determined to be a match. "," is determined to be a match.As a result, the character string organization means 55 creates a new character string consisting of the five characters "12345" (composition result), and this compilation result is sent as is. Used as credit data.

Here, a case will be described in which the character recognition device 50 having the above configuration is applied to a money exchange data communication device.

FIG. 7 is a block diagram of a money exchange data communication device according to the present invention.

In the exchange data communication device 20 shown in the figure, the branch 2
1 includes a character recognition device 50, an optical character reader (OCR) 22, a keyboard 23, and a network control unit (NCU) 24 connected to the character recognition device 50. District center 2
6 is a character recognition device 50 and a network control unit connected to this character recognition device 50 (N
CU) 28 and terminal devices 29a and 29b. The NCU 24 of the branch 21 is connected to the NCU 28 of the district center 26 via the public communication network 30.

The character recognition device 50 of the district center 26 is connected to the dedicated communication line 31.
It is connected to the host device 32 via.

The OCR 22 of the branch 21 optically reads the characters written on the exchange form 1 and outputs the image data. The keyboard 23 is for inputting kana and numerical values. The NCU 24 is composed of a modem and the like that converts data within the branch 21 into an audible signal for transmission to the public communication network 30.

The character recognition device 50 of the district center 26 recognizes characters from the data (image data) of the exchange form 1 received from the branch 21. Like the NCU 24, the NCU 28 is composed of a modem and the like. Terminal equipment 29
a, 29b are used to confirm character data etc. recognized by the character recognition device 50 and to instruct transmission to the host device 32.

Note that the character recognition device 50 provided in the business office 21 recognizes characters other than Kanji. Further, it is assumed that the character recognition device 50 provided at the district center 26 recognizes Chinese characters. In this way, by limiting the characters to be recognized, the branch office 21 can install a relatively simple character recognition device 50, and the district center 2
6 can be transmitted to less data.

Now, the operation of the money exchange data communication device 20 having the above configuration will be explained with reference to FIG.

FIG. 8 is an operational flowchart of the exchange data communication device 20 according to the present invention.

In the figure, first, the exchange form 1 is set in the 0CR22 of the sales branch 21, and the character data is read (step 5IO).The character data read by the 0CR22 is input to the character recognition device 50 as image data. Ru. This character recognition device 50 performs recognition and organization (character string organization processing) of the image data other than kanji characters (numeric values, kana characters, etc.) (step 5ll).

To explain this character string organization process in detail, taking the case of the necessary characters in the character string, for example item number 3 explained earlier in FIG. “12.345” with reference to areas 57a and 57b
Unnecessary characters before and after , “blank (b)”, “\”, “-
” etc. are deleted based on the comparison result of the comparison means 54 (
step 5100), then the first and last required characters (1
and 5) with the unnecessary characters stored in the unnecessary character storage section 57. This comparison is performed sequentially from the beginning of the required characters. First, it is determined whether there are any character strings left to be compared (or not) (step 3101). If the result is NO1, that is, a character string exists, one character is extracted from the character string (step 3101). Step 5102), and the comparison means 53 refers to the third area 57c of the unnecessary character storage means 57 and compares whether or not it is an unnecessary character (step 5103).
If so, storage is performed by the character string organizing means 54 (step 5I04), and the process returns to step 5IOI. Thereafter, the loop from step Sl to step 5104 is repeated until the comparison means 53 finishes comparing all character strings (consisting of necessary character strings). That is, 12.34
In the case of 5", it is the end of 6 comparisons from 1' to "2"..."5". During this time, unnecessary characters "," are deleted. One item is completed. Then, the result of step 8100 becomes YES, and the process moves to step 5105. In step 5105, it is determined whether recognition has been completed for all entry fields of exchange form l. If the result is NO, the process returns to step 5too. , in the case of YES, the process ends.

Next to step Sll, which is carried out according to the procedure described above, the recognition result is transmitted to the district center 26 (step 512), and the processing on the branch 21 side is ended.

Now, the district center 26 receives character data from the branch 21 (step 520). This data is composed of image data of kanji characters and character codes of other characters. The character recognition means 50 of the district center 26 temporarily stores the data received from the branch 21 in a storage element (RAM), etc. (not shown) (step S21), and the character recognition means 50 extracts character data to be recognized from the received data. Extract (image data) (step 522)
. In this case, image data of kanji characters is extracted. Then, the character recognition device 50 performs character string organization processing to obtain a new character string (step 523). This character string organization process is performed in steps 5100 to 3105 described earlier.
This is the process. Thereafter, data corresponding to each item of the exchange form 1 is displayed on the display (CRT) of the terminal device 29a (step 524), and the operator makes corrections to the displayed data as necessary (step 525).

Then, a send key (not shown) provided on the terminal device 29a is pressed (step 526). The character recognition device 50 assembles a message to be sent to the host device 32 (step 527), and sends the data to the host device 32 via the dedicated communication line 31 (step 828).

As explained above, the character recognition device 50 of the present invention can recognize characters before and after a necessary character from a character string, and (
Since unnecessary characters included in the middle) are deleted, only the necessary characters can be reliably obtained in any arrangement.

The character recognition device of the present invention is not limited to the above embodiments.

First to third areas 57a, 5 of unnecessary character storage means 57
The contents of the unnecessary characters stored in 7b and 57c may be the same or not the same, and there may be an area in which no unnecessary characters are stored. Further, in the embodiment, explanation has been given using the exchange data communication device 20, but it can be applied to any device that needs to extract only necessary characters from a string containing unnecessary characters. be able to.

Furthermore, the data to be recognized is not necessarily data from the OCR22 or the like, but may be a signal output from any device as long as it is image data.

(Effects of the Invention) The character recognition device of the present invention configured as described above can reliably recognize and delete unnecessary characters from a character string, so only necessary characters can be reliably recognized without being influenced by the arrangement of characters. It can be taken out. For this reason, when writing characters on forms, etc., there is no need to limit the writing method to a specific writing method, and there is no need to prepare and memorize many examples of writing in advance. There is no need to prepare a storage device with a large capacity.

[Brief explanation of the drawing]

Fig. 1 is a block diagram showing an embodiment of the character recognition device of the present invention, Fig. 2 is a surface diagram of a money order slip, Fig. 3 is a block diagram of a conventional money order data communication device, and Fig. 4 is a conventional character recognition device. 5 is an explanatory diagram of the operation of the conventional character recognition device, FIG. 6 is an explanatory diagram of the operation of the character recognition device of the present invention, and FIG. 7 is a diagram of the exchange data communication device according to the present invention. A block diagram of the data communication device, FIG. 8 is an operation flowchart of the money exchange data communication device according to the present invention. 51... Processor, 52... Storage device, 53...
- Recognition means, 54... Comparison means, 55... Character string organization means, 56... Character pattern storage means, 57... Unnecessary character storage means,

Claims (1)

[Scope of Claims] Recognition means for recognizing a character string consisting of two or more characters; unnecessary character storage means for storing unnecessary characters that do not require final character recognition; each character constituting the character string and the A character recognition device comprising: a comparison means for comparing the unnecessary characters with unnecessary characters; and a character string formation means for deleting the unnecessary characters from the character string to obtain a new character string based on the comparison result of the comparison means. .