JP7089991B2 - Office center operation support system and office center operation support method - Google Patents

Description

The present invention relates to an office center operation support system and an office center operation support method.

So-called high-performance ATMs that can process various payment slips such as tax and public money have appeared. According to the high-performance ATM, the user can process various payments, which were conventionally limited to the counter. Therefore, it is expected that the congestion at the counters in the branch offices will be eliminated and the convenience of users will be improved.

As a characteristic function of the above-mentioned high-performance ATM, there is a form recognition technique. The technology extracts and recognizes necessary information from the entire image of the form scanned by the scanner, and at the time of issuing the receipt, it is possible to issue a receipt printed with the form image and the receipt stamp.

As a conventional technique related to such a recognition technique, for example, a character recognition device that recognizes characters in an image, cuts out a local image while scanning an input image, extracts a feature, and registers the feature and a dictionary. For the first-stage character recognition means that determines character candidates based on the degree of similarity to the characteristics of the determined characters, and for the determined character candidates, characteristics different from the characteristics already used for feature extraction are extracted, and the characteristics are extracted. A character recognition device (see Patent Document 1) characterized by having one or more different characteristic character recognition means for narrowing down the character candidates based on the similarity with the characteristics of the characters registered in the dictionary has been proposed. Has been done.

In addition, the counter terminal device of the financial institution that processes the temporary receipt payment of the form in which the customer fills in the necessary items including the transaction amount by hand, and receives the information of the temporary receipt payment processing to correspond to the reception serial number and the transaction amount. The handwritten transaction amount is based on the host computer that issues the authentication amount to be issued and then returns to the counter terminal device, and the form reading data in which the reception serial number and the authentication amount are printed on the counter terminal device and converted into OCR. A form data processing system (see Patent Document 2), which is characterized by including a collation means installed in a centralized processing center for collating with the certified amount, has also been proposed.

In addition, a form type file in which keyword words for determining the type of form are registered in advance and image data of the form are read, a character string is extracted from the read image data, and character recognition is performed for the extracted character string. The section, the form type discrimination keyword word extraction section that extracts the character string recognized by the character recognition section as a keyword word, and the extracted keyword word and the form type discrimination keyword word registered in the form type file. A form type discriminating device (see Patent Document 3), which comprises a discriminating means for discriminating a form type by collating, has also been proposed.

Japanese Unexamined Patent Publication No. 2006-251920 Japanese Unexamined Patent Publication No. 2006-4121 Japanese Unexamined Patent Publication No. 200-

By the way, at a financial institution that has introduced the above-mentioned high-performance ATM, as a closing work for transactions at the high-performance ATM, the bank employees collate and confirm the physical form and the transaction result for each branch. It is being carried out.

On the other hand, there is also a need to consolidate this tightening work at the office center instead of performing it at each branch. Therefore, if the closing work of each branch is to be centralized at the office center, depending on the financial institution, an enormous amount of man-hours of about a dozen man-days is required, which is unrealistic in terms of operating costs. On the other hand, even if a recognition technique such as the conventional technique is applied, it is difficult to automate the above-mentioned tightening work, and it cannot be an effective option.
Therefore, an object of the present invention is to provide a technique for making a huge amount of tightening work in an office center efficient.

The office center operation support system of the present invention that solves the above problems is based on the identification information of each form used in the financial institution, the form information describing the attribute information associated with the identification information, and the predetermined transaction in the financial institution. The storage unit that holds the image data of the used form and the identification information of the transaction specified by the predetermined recognition process for the image data of the form, and the recognition process obtained for the form of the transaction to be closed. The first method used in the recognition process in the target transaction and the extraction process of acquiring the identification information specified in the above from the storage unit and extracting the attribute information of each form in the form group corresponding to the identification information from the form information. The second recognition engine, which is different from the recognition engine of the above, executes the recognition process for the image data of the form of the target transaction to specify the attribute information of the form, and the attribute extracted by the extraction process. It is characterized by including a calculation unit that collates the information with the attribute information specified by the specific process and executes a predetermined completion process related to the closing process for the target transaction in which the attribute information matches.

Further, the office center operation support method of the present invention is used in the identification information of each form used in the financial institution, the form information describing the attribute information associated with the identification information, and the predetermined transaction in the financial institution. An information processing system provided with a storage unit that holds the image data of the form and the identification information of the transaction specified by the predetermined recognition process for the image data of the form is obtained with respect to the form of the transaction to be closed. The identification information specified in the recognition process is acquired from the storage unit, and the attribute information of each form in the form group corresponding to the identification information is extracted from the form information. The second recognition engine, which is different from the first recognition engine, executes the recognition process for the image data of the target transaction form, and specifies the attribute information of the form, and extracts by the extraction process . It is characterized in that the attribute information specified is collated with the attribute information specified by the specific process, and a predetermined completion process related to the closing process is executed for the target transaction in which the attribute information matches.

According to the present invention, a huge amount of tightening work in an office center can be made efficient.

It is a network configuration diagram including the office center operation support system of this embodiment. It is a figure which shows the hardware configuration example of the tax public money server in this embodiment. It is a figure which shows the hardware configuration example of the high-performance ATM in this embodiment. It is a figure which shows the data structure example of the form specific knowledge of this embodiment. It is a figure which shows the data structure example of the form image data of this embodiment. It is a figure which shows the data structure example of the transaction identification information of this embodiment. It is a figure which shows the data structure example of the acceptance / rejection knowledge of this embodiment. It is a figure which shows the flow example of the office center operation support method in this embodiment. It is a figure which shows the screen example in this embodiment.

－－－ Network configuration －－－

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a network configuration diagram including the office center operation support system 10 of the present embodiment. The office center operation support system 10 shown in FIG. 1 is a computer system that makes enormous tightening work in the office center efficient.

The office center operation support system 10 in the present embodiment is a system operated by a financial institution, and as shown in FIG. 1, is installed in a tax public money server 100 operated by the headquarters of the financial institution and each branch office. It is composed of a high-performance ATM 200. The tax public money server 100 and the high-performance ATM 200 are connected to each other so as to be able to communicate with each other via a network 1 inside a secure financial institution.

Of course, such a configuration is an example, and either the tax / public money server 100 or the high-performance ATM 200 may have the configuration / function as the office center operation support system 10.

Of the above configurations, the tax public money server 100 cooperates with each of the high-performance ATMs 200 provided in each business office, and provides the customer with a tax public money payment transaction service using a form as a medium at each business office.

More specifically, it is a server device that determines the subsequent operation contents according to the type of the form received from the customer in the high-performance ATM 200 and instructs the high-performance ATM 200 of the operation contents. As the above-mentioned operation contents, it is assumed whether or not the transaction can be accepted.

The tax public money server 100 of this embodiment has a function of recognizing various forms. This recognition function extracts and recognizes necessary information from the form image obtained from the high-performance ATM 200.

In addition, the high-performance ATM 200 is an ATM that can handle various payment slips (hereinafter referred to as forms) such as tax and public money of tens of thousands of types, which were conventionally limited to handling at the counter.
From the customer who is a user of the high-performance ATM 200, it is not necessary to fill in the payment request form, and various payment processes can be performed by his / her own operation. As a result, it tends to lead to improvement of customer convenience and customer satisfaction.

The high-performance ATM 200 of the present embodiment has a function of recognizing various forms, although the feature extraction algorithm is different from that of the tax and public money server 100. This recognition function extracts and recognizes necessary information from the form image read by the scanner unit, and notifies the tax and public money server 100 of the result. Unlike conventional devices and the like, form recognition is possible without format registration such as predetermining the area to be read. In addition, when issuing a receipt when the transaction is completed, the receipt is issued by printing the receipt stamp on the form image, and it is not necessary to stamp the original payment slip at the counter.
--- Hardware configuration ---

Next, the hardware configuration of the tax and public money server 100 constituting the office center operation support system 10 of the present embodiment will be shown. FIG. 2 is a diagram showing a hardware configuration example of the tax and public money server 100 of the present embodiment.

In this case, the tax and public money server 100 is used in a storage unit 101 composed of an appropriate non-volatile storage element such as an SSD (Solid State Drive) or a hard disk drive, a memory 103 composed of a volatile storage element such as a RAM, and a storage unit 101. The program 102 to be held is executed by reading it into the memory 103 to perform overall control of the device itself, and is connected to a calculation unit 104 such as a CPU that performs various determinations, calculations and control processes, and network 1 to be connected to the height of each sales office. Function A communication unit 105, which is responsible for communication processing with each of the ATMs 200, is provided.

In addition to the general information processing function, the program 102 held by the storage unit 101 has a function required as a device constituting the office center operation support system 10 of the present embodiment, that is, a function of recognizing various forms. It is a program which implements the recognition engine 111 of 2.

Further, the storage unit 101 stores at least the form specific knowledge 125, the form image data 126, the transaction identification information 127, and the acceptance / rejection knowledge 128. Details of these information 125 to 128 will be described later.

Further, FIG. 3 shows an example of the hardware configuration of the high-performance ATM 200 of the present embodiment. The high-performance ATM 200 of the present embodiment includes a storage unit 201 composed of an appropriate non-volatile storage element such as an SSD (Solid State Drive) or a hard disk drive, a memory 203 composed of a volatile storage element such as a RAM, and a storage unit 201. The program 202 to be held is executed by reading it into the memory 203, etc., and the integrated control of the device itself is performed. Section 205, an output section 206 such as a display that displays processing data, a communication section 207 that is connected to network 1 and is responsible for communication processing with the tax and public money server 100 described above, a scanner unit 208 that is responsible for image acquisition operations for forms, and , A print unit 209, which is responsible for the above-mentioned receipt issuing operation.

In addition, it is assumed that it naturally has a configuration / function such as a cash hangar and a discharge / receipt mechanism, and a cash transfer mechanism from the hangar to the discharge / receipt mechanism, which are common as ATMs (general). Not shown because it is an existing technology).

In addition to the general information processing function, the program 202 provided in the storage unit 201 has a function required as a device constituting the office center operation support system 10 of the present embodiment, that is, a twelfth function of recognizing various forms. It is a program that implements the recognition engine 110 of.

The first recognition engine 110 described above executes character recognition processing on the image data of the form obtained by the scanner unit 208, and acquires the keyword (attribute information) described in the form. However, the feature extraction algorithm used is different between the first recognition engine 110 and the second recognition engine 111.

For example, when the feature extraction algorithm adopted by the first recognition engine 110 is a so-called superposition method, the feature extraction algorithm adopted by the second recognition engine 111 is a so-called directional matching method. Imagine the situation.
Of course, the feature extraction algorithm in the recognition engine is not limited to these specific examples, and may be appropriately adopted.

The storage unit 201 of the high-performance ATM 200 may also hold the form specific knowledge 125, the form image data 126, the transaction identification information 127, and the acceptance / rejection knowledge 128, as in the tax public money server 100. ..
--- Data structure example ---

Subsequently, the tables used by the office center operation support system 100 of the present embodiment will be described. FIG. 4 shows an example of the form-specific knowledge 125 in this embodiment.
The form-specific knowledge 125 of the present embodiment is a table that stores attribute information of each form used by a financial institution.
The data structure is a collection of records associated with keywords that are attribute information, with the handling type as the key.

Of these, the handling type is a value that is also used in the acceptance / rejection knowledge 128 described later, and is a value that is uniquely determined for each combination of keywords described in the form. It should be noted that each handling type is associated with a predetermined value as to whether or not it is specified as a handling target at the branch office of the above-mentioned financial institution in the acceptance / rejection knowledge 128.

In addition, the keywords are keywords written (including printed) on the form, and specifically, the names of places such as prefectures and municipalities, and the names of tax and public funds such as municipal tax and property tax are assumed.

Further, FIG. 5 shows an example of the form image data 126 in the present embodiment. The form image data 126 of the present embodiment is a table in which the data of the form image obtained by the high-performance ATM 100 by the scanner unit 108 is stored.
The data structure is a collection of records composed of data such as a storage destination of form image data and an image shooting date and time, using a transaction ID as a key.

Further, FIG. 6 shows an example of the transaction identification information 127 in the present embodiment. The transaction identification information 127 of the present embodiment is a table that stores information on the handling type specified by the tax and public money server 100 in the form specific knowledge 125 with respect to the form image data of each transaction in the above-mentioned form image data 126.
The data structure is a collection of records associated with the value of the handling type specified for the form of the transaction, with the transaction ID as the key.

Further, FIG. 7 shows an example of acceptance / rejection knowledge 128 in this embodiment. The acceptance / rejection knowledge 128 of this embodiment is a table that defines the conditions of the forms to be handled at the business office of the financial institution.
The data structure is a collection of records associated with values such as payment deadline, delinquent charge, and acceptance / rejection, using the handling type as a key.
Of these, the payment deadline stipulates that the payment deadline stated in the form indicated by the handling type must be before the lapse.
In addition, the delinquent charge stipulates the necessity that there is no delinquent charge stated in the form indicated by the handling type.
In addition, whether or not the form can be accepted is a value specified as to whether or not the above-mentioned business office accepts the form.
－－Flow example －－－

Hereinafter, the actual procedure of the office center operation support method in this embodiment will be described with reference to the figure. Various operations corresponding to the office center operation support method described below are realized by programs that are read and executed by the tax and public money server 100 and the high-performance ATM 200, which constitute the office center operation support system 10, respectively. .. The program is composed of codes for performing various operations described below.

FIG. 8 is a diagram showing a flow example of the office center operation support method in the present embodiment. Here, it is assumed that a customer operates a high-performance ATM 200 and performs a predetermined transaction procedure using a form at a branch office of a financial institution during normal business hours (example: until 15:00). ..
In this case, the high-performance ATM 200 captures the form brought by the customer with the scanner unit 208 and acquires the form image data (s10).

Further, the high-performance ATM 200 inputs the form image data obtained in the above-mentioned s10 into the first recognition engine 110, and specifies the content described in the form by character recognition (s11). What is specified here assumes a keyword.

Subsequently, the high-performance ATM 200 associates the form image data and keywords obtained in s10 and s11 with the transaction ID assigned to the transaction, and transmits the form image data and keywords to the tax and public money server 100 via the network 1 (s12).

On the other hand, the tax public money server 100 receives the form image data and the keyword associated with the transaction ID transmitted from the above-mentioned high-performance ATM 200 (s13).

Further, the tax public money server 100 generates a record using the transaction ID as a key for the form image data obtained in s13, and stores it in the form image data 126 (s14).

Subsequently, the tax and public money server 100 collates the keyword obtained in s13 with each record of the form specific knowledge 125, and handles the corresponding form (the one targeted by the high-performance ATM 200 in s10 for shooting the form image data). (Identification information of the form) is specified (s15).

For example, if the keywords obtained in s13 are "Hokkaido", "Sapporo City", and "Municipal Tax", the set of these keywords is collated with each record of the form specific knowledge 125 illustrated in FIG. The handling type "1111111111" can be specified.

Further, the tax public money server 100 generates a record in which the handling type specified in s15 is associated with the transaction ID obtained in s13, and stores it in the handling type information 127 (s16).

Subsequently, the tax and public money server 100 collates the handling type specified in s15 with each record of acceptance / rejection knowledge 128, identifies a record having a matching value in the "handling type" column, and determines the conditions included in the record. That is, it is determined whether or not the description of the form matches the payment deadline and the delinquent charge (s17).

For example, the tax public money server 100 has a payment deadline of "OK" and a delinquent charge of "no" based on the handling type "1111111111" of the target form, or the description of the corresponding form (keyword obtained in s13). Judgment is based on.
The tax and public money server 100 specifies whether or not the form can be accepted according to the result of the above determination, and notifies the high-performance ATM 200 of the specific result (s18).

For example, as a result of the above determination, if the description of the form matches with respect to the payment deadline and the delinquent charge, the tax public money server 100 notifies the high-performance ATM 200 that the form can be accepted. On the other hand, as a result of the above determination, if the description of the form does not match with respect to the payment deadline and the delinquent charge, the tax public money server 100 notifies the high-performance ATM 200 that the form cannot be accepted.

On the other hand, the high-performance ATM 200 receives the above-mentioned specific result of acceptance / rejection from the tax / public money server 100, and executes the business operation according to the specific result with respect to the corresponding form currently being processed (s19). When the specific result indicates "cannot be accepted", the high-performance ATM 200 displays a message such as "this form cannot be handled" on the output unit 206, and ends the process. When the specific result indicates "acceptable", the high-performance ATM 200 will execute a predetermined procedure according to the transaction.

The tax and public money server 100 and the high-performance ATM 200 will repeatedly execute each of the above-mentioned processes of s10 to s19 for each transaction involving a form until the business hours of the business office of the financial institution are completed.

Here, after business hours at the above-mentioned financial institution (eg, after 15:00), the tax and public money server 100 is used as a closing process at the office center, for example, at the high-performance ATM 200 of each branch office for the day. Transactions that involve the form processing of the above are the target transactions for closing processing, and the subsequent processing is performed.

In this case, the tax and public money server 100 identifies the transaction type obtained with respect to the form processed in the target transaction (specified by the transaction ID. For example, "t00000001") as the information of the target transaction of the above-mentioned closing process. A record group (attribute information of each form in the form group) corresponding to this handling type is extracted from the form specific knowledge 125 (s20), which is acquired from the information 127.

For example, if the handling type obtained for the form processed in the target transaction is "11111111111" as the information of the target transaction "t00000001" of the closing process, three records are shown as a record group corresponding to this handling type. It will be extracted from the form specific knowledge 125 exemplified in 4.

These three records include a record with keyword 1 being "Hokkaido", keyword 2 being "Sapporo city", keyword 3 being "municipal tax", keyword 1 being "Hokkaido", keyword 2 being "Sapporo city", and keyword 3 being. The record of "property tax", keyword 1 is "Hokkaido", keyword 2 is "Otaru city", and keyword 3 is "municipal tax".
Here, the record group extracted by the tax and public money server 100 is a population of characters to be recognized by the second recognition engine 111 used for the closing process.

Subsequently, the tax and public money server 100 acquires the form image data of the transaction "t00000001" to be closed based on the information of the "form image data storage destination" indicated by the form image data 126, using the transaction ID "t00000001" as a key. , Character recognition processing by the second recognition engine 111 is executed for this form image data (s21).

The second recognition engine 111 is a recognition engine having a different feature extraction algorithm from the first recognition engine 110 used by the high-performance ATM 200 during business hours.

Further, the tax and public money server 100 identifies the keyword (attribute information) which is the description content of the form from the form image data by the character recognition process of s21 described above (s22).

In this s22, it is assumed that the tax public money server 100 obtains the keywords "Hokkaido", "Sapporo city", and "municipal tax" from the form image data of the transaction ID "t00000001".

Next, the tax and public money server 100 collates the keyword specified by each record in the record group extracted in s20 with the keyword specified in s22 (s23).

The records extracted in s20 above include a record in which keyword 1 is "Hokkaido", keyword 2 is "Sapporo city", keyword 3 is "municipal tax", keyword 1 is "Hokkaido", and keyword 2 is "Sapporo city". Keyword 3 was a record of "fixed asset tax", keyword 1 was "Hokkaido", keyword 2 was "Otaru city", and keyword 3 was a record of "municipal tax". It is possible to identify one record including the keywords "Hokkaido", "Sapporo city", and "municipal tax" obtained from the transaction form image data of the transaction ID "t00000001".

As a result of the above collation, when the keyword specified by each record in the record group extracted in s20 does not match the keyword specified in s22 (s24: n), the tax and public money server 100 is the transaction subject to the closing. (See screen 1000 in FIG. 9) is notified to the terminal such as the person in charge of the closing process at the office center (s25), and the process is terminated.

On the other hand, as a result of the above-mentioned collation, when the keyword specified by any of the records in the record group extracted in s20 matches the keyword specified in s22 (s24: y), the tax and public money server 100 performs the matching. For the transaction to be closed, the predetermined completion process related to the closing process (eg, in the DB for transaction management, etc., the status value of the closing completion is given to the target transaction) is executed (s26), and the process is performed. To finish.

Although the best mode for carrying out the present invention has been specifically described above, the present invention is not limited to this, and various modifications can be made without departing from the gist thereof.
According to this embodiment, a huge amount of tightening work in the office center can be made efficient.

In the office center operation support technique of the present embodiment, the form presented by the customer is processed by the high-performance ATM during the normal business hours of the financial institution (eg, until 15:00), so that the customer's desired transaction can be performed. Imagine a situation where you are in the process of proceeding with. Regarding transactions via forms during this business hours, each operation of image data acquisition at a high-performance ATM for the form and character recognition (by the first recognition engine) from the image data is executed, and the form is displayed. The description content of is recognized. The high-performance ATM or the tax public money server collates the keywords included in the description contents recognized as described above with the above-mentioned form-specific knowledge to specify the form handling type (form identification information). In addition, the tax / public money server or the high-performance ATM collates the specified handling type with the acceptance / rejection knowledge to determine and execute the business operation of the form. The handling type and image data of each form obtained by the above-mentioned series of procedures are held at least in the above-mentioned tax and public money server.

On the other hand, after business hours at the above-mentioned financial institution (eg, after 15:00), as a closing process at the office center, the tax and public money server, for example, a transaction involving form processing at a high-performance ATM for the day. , Is the target transaction of the closing process, the handling type obtained for the corresponding form is acquired from the storage unit, and the record group (attribute information of each form in the form group) corresponding to this handling type is extracted from the form specific knowledge. .. The record group extracted here is a population of characters recognition targets by the second recognition engine used for the closing process.

Therefore, the tax and public money server executes the recognition process for the image data of the form of the transaction to be closed by the second recognition engine different from the first recognition engine used in the above-mentioned business hours, and the form is concerned. Specify the keyword (attribute information) included in the description of. In addition, the tax and public money server collates the keyword specified by each record in the above-mentioned record group with the keyword specified for the form of the transaction to be closed, and the closing target matches any of the records in the record group with the keyword. For the transaction of, the predetermined completion process related to the closing process (eg, in the DB for transaction management, etc., the status value of the closing completion is given to the target transaction) is executed.

The description herein reveals at least the following: That is, in the office center operation support system of the present embodiment, the calculation unit identifies, as a result of the collation, the target transaction whose attribute information does not match any of the forms in the form group as a confirmation target by personnel. It may be said that the process of outputting the information about the target transaction to the predetermined device is further executed.

According to this, unlike the transaction in which the closing process is completed by the series of processes up to the above-mentioned collation, the transaction requires confirmation work by personnel (example: the description content of the form is erroneously recognized by the recognition engine. As a result, it is possible to efficiently identify transactions that have been made with incorrect content, and only such transactions can be targeted for personnel confirmation work (at the office center). As a result, the enormous tightening work in the office center can be made more efficient.

Further, in the office center operation support system of the present embodiment, the calculation unit uses a second recognition engine having a feature extraction algorithm different from that of the first recognition engine to recognize the image data of the target transaction form. May be executed.

According to this, for example, even if different character recognition software is adopted as the first and second recognition engines, the difference between the two feature extraction algorithms is not large, and the characters for the image data are displayed from different aspects. By recognizing, the accuracy of tightening work can be further improved. As a result, the enormous tightening work in the office center can be made more efficient.

Further, in the office center operation support method of the present embodiment, the information processing system identifies, as a result of the collation, a target transaction whose attribute information does not match any of the forms in the form group as a confirmation target by personnel. Further, the process of outputting the information regarding the target transaction to the predetermined device may be executed.

Further, in the office center operation support method of the present embodiment, the information processing system recognizes the image data of the target transaction form by the second recognition engine having a feature extraction algorithm different from that of the first recognition engine. The process may be executed.

1 Network 10 Office center operation support system 100 Tax public money server 101 Storage unit 102 Program 103 Memory 104 Calculation unit 105 Communication unit 125 Form specific knowledge (form information)
126 Form image data 127 Transaction identification information 128 Acceptability Knowledge 200 High-performance ATM
201 Storage unit 202 Program 203 Memory 204 Calculation unit 205 Input unit 206 Output unit 207 Communication unit 208 Scanner unit 209 Print unit 225 Form specific knowledge (form information)
226 Form image data 227 Transaction identification information 228 Acceptability knowledge

Claims (6)

For the form information describing the identification information of each form used in the financial institution and the attribute information associated with the identification information, the image data of the form used in the predetermined transaction at the financial institution, and the image data of the form. A storage unit that holds the identification information of the transaction specified by the predetermined recognition process, and
An extraction process in which the identification information specified in the recognition process obtained for the form of the target transaction of the closing process is acquired from the storage unit, and the attribute information of each form in the form group corresponding to the identification information is extracted from the form information. And, the recognition process for the image data of the form of the target transaction is executed by the second recognition engine different from the first recognition engine used in the recognition process at the time of the target transaction, and the attribute information of the form is obtained. The specific process to be specified is collated with the attribute information extracted by the extraction process and the attribute information specified by the specific process, and a predetermined completion process related to the closing process is executed for the target transaction in which the attribute information matches. And the arithmetic unit to do
An office center operation support system characterized by including. The arithmetic unit
As a result of the collation, a target transaction whose attribute information does not match any of the forms in the form group is specified as a confirmation target by personnel, and a process of outputting information on the target transaction to a predetermined device is further executed. ,
The office center operation support system according to claim 1, characterized in that. The arithmetic unit
A second recognition engine having a different feature extraction algorithm from the first recognition engine executes recognition processing for the image data of the target transaction form.
The office center operation support system according to claim 1, characterized in that. For the form information describing the identification information of each form used in the financial institution and the attribute information associated with the identification information, the image data of the form used in the predetermined transaction at the financial institution, and the image data of the form. An information processing system equipped with a storage unit that holds the identification information of the transaction specified by the predetermined recognition process
An extraction process in which the identification information specified in the recognition process obtained for the form of the transaction subject to the closing process is acquired from the storage unit, and the attribute information of each form in the form group corresponding to the identification information is extracted from the form information. When,
A second recognition engine different from the first recognition engine used in the recognition process at the time of the target transaction executes the recognition process for the image data of the form of the target transaction to specify the attribute information of the form. Specific processing and
By collating the attribute information extracted by the extraction process with the attribute information specified by the specific process, for the target transaction in which the attribute information matches, the predetermined completion process related to the closing process and the predetermined completion process.
An office center operation support method characterized by executing. The information processing system
As a result of the collation, a target transaction whose attribute information does not match any of the forms in the form group is specified as a confirmation target by personnel, and a process of outputting information on the target transaction to a predetermined device is further executed.
The office center operation support method according to claim 4, characterized in that. The information processing system
A second recognition engine having a different feature extraction algorithm from the first recognition engine executes recognition processing for the image data of the target transaction form.
The office center operation support method according to claim 4, characterized in that.

Images