JP7199928B2 - CASH CENTER MONITORING SYSTEM AND METHOD - Google Patents

Description

The present invention relates to a cash center monitoring system and method.

Banknotes collected from an ATM (Automated Teller Machine) or the like are carried into a cash center and sorted by an automatic cash sorter (sorter) installed in the cash center (Patent Document 1). A typical automatic cash sorting machine is often used offline without being connected to a communication network, and most of them are not remotely monitored.

JP 2018-045631 A

Since conventional cash centers are not remotely monitored, even if banknotes are stolen by an employee, the fact of the theft is not known or it takes a long time to be discovered. Therefore, an employee who arbitrarily withdraws damaged banknotes or the like can retire from the cash center before the fraud is known.

On the other hand, in many countries, when counterfeit bills are detected at cash centers, they are obliged to report the counterfeit bills to the police and the central bank. There is also

SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide a monitoring system and method for a cash center that can improve reliability and usability.

In order to solve the above problems, a monitoring system for a cash center according to the present invention comprises a monitoring camera for photographing an operator who operates an automatic cash sorting machine, and a management device communicably connected to the automatic cash sorting machine and the monitoring camera. If a preset event is detected, the image data of the operator corresponding to the event is extracted from the video file captured by the surveillance camera, and sent from the automatic cash sorting machine. and a management device for associating sorting result data corresponding to a predetermined event among the sorting result data with the extracted image data and providing the data to a predetermined output destination.

According to the cash center monitoring system of the present invention, when a predetermined event is detected by monitoring the automatic cash sorting machine from the management device, the image data of the operator corresponding to the predetermined event is extracted, and a corresponding action is taken. The data can be output to a predetermined output destination in association with the sorting result data. Furthermore, since the image data and sorting result data can be used easily, usability is improved.

1 is a functional configuration diagram showing the entire cash center monitoring system; FIG. It is an overall explanatory diagram showing the relationship between ATM, sorter, and management server. It is an example of a transaction management database. It is an example of a serial number list. This is an example of a database that manages videos of fraud patterns. This is an example of a database that manages surveillance videos. This is an example of a database that manages employees. 10 is a flow chart showing processing for registering a video of a fraud pattern. 4 is a flow chart showing processing for recording an operator's monitoring video. FIG. 11 is a flow chart showing a process of outputting evidence of fraud; FIG. It is a flow chart which shows automatic cash sorting processing of a sorter. FIG. 11 is a flowchart showing processing for registering sorter sorting results; FIG. 10 is a flow chart showing processing for outputting evidence of counterfeit bills. It is a flowchart which shows the process which outputs the evidence of amount-of-money discrepancy. Fig. 10 is a flow chart showing a process for printing evidence or copying it to an external memory;

BEST MODE FOR CARRYING OUT THE INVENTION An embodiment of the present invention will be described below based on the drawings. In the monitoring system 3 according to this embodiment, the sorter 20 is brought online, and the management server 1 interlocked with the monitoring camera 21 and the sorter 20 are connected by the communication network CN. As a result, in the monitoring system 3 of the present embodiment, all of the banknote information sorted by the sorter 20, monitoring images (monitoring videos) captured by the monitoring camera 21, and analysis results of the monitoring images are sent to the management server 1. Can be centrally managed. As a result, it is possible to efficiently perform processes such as detection of fraud, discrepancies in amounts, collection of evidence, and reporting of detection of counterfeit bills, thereby improving reliability and usability.

The cash center monitoring system 3 according to this embodiment can remotely monitor an operator 22 who operates a sorter 20 in a cash center 2 that sorts banknotes collected from bank branches or ATMs 4 . A monitoring camera 21 installed in the cash center 2 is connected to the management server 1 . The management server 1 compares the motion of the operator with a pre-registered moving image of a fraudulent behavior pattern to determine whether fraudulent behavior has occurred.

The sorter 20 stores, for example, a transaction ID (identification information) for distinguishing the unit of banknotes to be sorted, the number of sorted banknotes and denominations, the fitness status, the serial number of the banknotes, and the sorting date and time. is recorded as banknote information in the storage device. The sorter 20 also records the collected banknotes of which branch number, which branch name, or from which ATM number. Further, the sorter 20 stores the login ID and login password of the operator who sorted the banknotes, the sorting start time and sorting end time, the number of the sorter that sorted the banknotes, the recorded banknote information, and the information on the collection source of the banknotes. and operator information to the management server 1 .

When an operator who has committed a fraudulent act is detected, the management server 1 immediately identifies the operator and extracts video files that are highly likely to include fraudulent activity. The management server 1 associates personal identification information such as the name of the operator who committed the fraudulent act, bibliographic information (date and time, amount of money, etc.) indicating the content of the fraudulent act, and image data as evidence of the fraudulent act. In addition to saving, the information (predetermined data) is transmitted to the terminal 5 used by the administrator 51 . The image data may be sent directly to the administrator terminal 5 or may be provided to the administrator terminal 5 when the administrator terminal 5 accesses the management server 1 .

Furthermore, even when the result of sorting the banknote collection cassettes 41 taken out from the ATM 4 by the sorter 20 and the amount of money on the receipt 42 output from the ATM 4 do not match, the management server 1 causes the management server 1 to transmit the result to the terminal 5 of the manager 51 . can be done.

Furthermore, when a counterfeit bill is detected by the sorter 20 , the management server 1 can associate the image data of the counterfeit bill with the bibliographic information and transmit them to the administrator terminal 5 .

According to this embodiment configured in this way, the management server 1 can quickly detect the discovery of a counterfeit bill and the fraud (theft of banknotes) inside the cash center 2, and can prevent such fraudulent activities. Evidence (who, when, what type of sorting transaction occurred, etc.) can be automatically output. Therefore, the reliability of the cash center can be improved, and the usability can be improved.

A first embodiment will be described with reference to FIGS. 1 to 15. FIG. FIG. 1 is a functional configuration diagram showing the entire cash center monitoring system 3. As shown in FIG.

The monitoring system 3 includes, for example, a management server 1, a cash center 2, and an administrator terminal 5. FIG.

First, the cash center 2 will be explained. The cash center 2 sorts banknotes collected from the ATM 4 or the like and reports the sorting result to the management server 1 . Below, the case where the banknotes collect|recovered from ATM4 are sorted is mentioned as an example, and is demonstrated.

A plurality of banknote recovery cassettes (hereinafter referred to as cassettes) 41 and receipts 42 are carried into the cash center 2 from the ATM 4 . The cassette 41 is prepared according to the purpose, for example, a cassette for 10,000 yen bills, a cassette for 5,000 yen bills, a collection cassette, or the like. The receipt 42 is printed with, for example, the total amount of collected banknotes, the name of the branch, the ATM number, the date of collection, and the like. The receipt 42 is a recording medium that holds information on banknotes input to the sorter 20 for sorting processing, and can also be called cash information for sorting.

The sorter 20 includes a plurality of insertion slots 201 into which banknotes from the cassettes 41 are inserted, and a user interface device 202 for operating the sorter 20 . The user interface device 202 can be configured, for example, like a tablet-type computer terminal having an information input section and an information output section. A user interface device other than a tablet computer terminal may be used.

The operator 22 wears, for example, a bib 221 on which the operator ID is written on the front and back of the clothes. The operator 22 may wear a hat (not shown) displaying the operator ID. By conspicuously displaying the operator ID of the operator, the operator 22 can be easily identified from the moving image file captured by the surveillance camera 21 . Instead of the number 221, information of an information terminal (a so-called smart phone, an IC tag, etc.) worn by the operator 22 can also be used. In this case, a reader for contactless reading of the operator ID from the information terminal may be arranged at or near the sorter 20 . However, using the number 221 can be operated at a lower cost.

A surveillance camera 21 is installed on the ceiling of the cash center 2 or the like. The mounting position and angle of the monitoring camera 21 are determined so that the action of the operator 22 operating the sorter 20 can be photographed. A monitor camera 21 may be arranged for each sorter 20, or if possible, a plurality of sorters 20 may be monitored by a single monitor camera 21. FIG. Alternatively, a plurality of monitoring cameras 21 may be arranged for one sorter 20 . A video (moving image) captured by the monitoring camera 21 is transmitted to the management server 1 from the communication network CN.

The administrator terminal 5 is an information terminal used by the administrator 51 . A so-called smart phone can be used as the administrator terminal 5 . The administrator terminal 5 is not limited to a smart phone as long as it has a function of communicating with the management server 1 and a function of exchanging information with the management server 1 . The manager terminal 5 receives from the management server 1 the information about the fraud that occurred at the cash center 2 and the information about the counterfeit bills detected by the sorter 20 , and provides the manager 51 with the information. E-mail, for example, can be used to provide information from the management server 1 to the administrator terminal 5 . Information may be exchanged between the management server 1 and the administrator terminal 5 in a form other than e-mail.

A configuration of the management server 1 will be described. The management server 1 is configured as a computer including circuits such as a microprocessor, a main memory device, and an auxiliary memory device (none of which are shown). A management server 1 serving as a management device for remotely monitoring the cash center 2 is implemented by the microprocessor executing a predetermined computer program read into the main memory.

The management server 1 includes, for example, a fraud detection unit 10, a database 11, a communication unit 12, a sorting result data acquisition unit 13, a video data acquisition unit 14, an output unit 15, and a user interface unit (UI ) 16 and a memory interface unit 17 .

The fraud detection unit 10 has a function of detecting a predetermined event registered in advance. The function of detecting a predetermined event includes, for example, a fraudulent action detection unit 101 that detects actions of the operator 22 that are similar to previously registered fraudulent action actions, and the contents of the receipts 42 and the sorting result of the sorter 20 . and a counterfeit bill detection unit 103 for detecting that counterfeit bills are found among the bills input to the sorter 20 .

Furthermore, the fraud detection unit 10 includes an evidence output unit 104 that outputs evidence in response to a request from the administrator terminal 5 . Predetermined data output from evidence output unit 104 is sent from output unit 15 to administrator terminal 5 via communication unit 12 and communication network CN. Note that the evidence output unit 104 can be provided as a function separate from the fraud detection unit 10 .

The database 11 manages information used for detection of fraud and reporting to the administrator 51 of detected fraud. Examples of information stored in the database 11 will be described later with reference to FIGS. 3 to 7. FIG.

The communication unit 12 has a function of communicating with the sorter 20, the monitoring camera 21, and the administrator terminal 5 via the communication network CN. The communication network CN will be described later with reference to FIG.

The sorting result data acquisition unit 13 has a function of acquiring data from the sorter 20 via the communication network CN and the communication unit 12 . The sorting result data acquisition unit 13 can also acquire information written on the receipt 42 via the sorter 20 . The acquired data is stored in the database 11 .

The moving image data acquisition unit 14 has a function of acquiring a moving image file captured by the monitoring camera 21 via the communication network CN and the communication unit 12 . The acquired moving image file is stored in the database 11 .

The output unit 15, which can also be called an "evidence providing unit", transmits the facts of fraud detected by the fraud detection unit 10 and evidence thereof from the communication unit 12 to the administrator terminal 5 via the communication network CN. It is a function to E-mail, for example, can be used to transmit information (data). The output unit 15 may have a web server function and provide information to the administrator terminal 5 in the form of web pages.

The user interface unit 16 is an information input/output device used for maintenance of the management server 1 and the like. The user interface unit 16 can include, for example, a keyboard, pointing device, touch panel, display, speech synthesizer, printer, and the like.

The memory interface unit 17 is an interface that transmits and receives data to and from the external memory 6 . The evidence (predetermined data) output by the evidence output unit 104 can be written from the memory interface unit 17 to the external memory 6 . Alternatively, the evidence output by the evidence output unit 104 can be output to the printer or display of the user interface unit 16 .

FIG. 2 is a diagram showing the overall connection configuration. FIG. 1 shows the case where the management server 1, the sorter 20, the monitoring camera 21, and the administrator terminal 5 are connected by a common communication network CN. Instead of this, the management server 1 and each sorter 20 are connected by a communication network CN1, the management server 1 and each monitoring camera 21 are connected by another communication network CN2, and the management server 1 and the administrator terminal 5 are connected. Furthermore, it may be connected by another communication network CN3.

The management server 1 can monitor multiple sorters 20 . A plurality of sorters 20 may be provided in the same cash center 2 or may be provided in different cash centers 2 . If the management server 1 is configured as a so-called cloud system, it is possible to provide monitoring services to different businesses (banks).

FIG. 3 shows an example of the transaction management database T10 included in the database 11 of the management server 1. As shown in FIG.

The transaction ID C101 is a number for identifying which group of banknotes has been sorted when the banknotes to be collected in the cash center 2 are sorted by group. For example, sorting all bills collected from one ATM 4 is regarded as one transaction, and a transaction ID is given.

The transaction start time C102 is the time at which the sorting of one transaction (one transaction) is started, that is, when the operator 22 presses a transaction start button (not shown) on the screen of the user interface device 202 provided in the sorter 20. is the time recorded in The transaction end time C103 is the time when the sorting of one transaction is completed, that is, the time recorded when the operator 22 presses the transaction end button (not shown).

The operator name/ID C104 is an employee number that identifies the operator 22 . The sorter number C105 is a number that identifies each sorter 20 . The collection date C106 is the date on which the banknote was collected from the branch office or the ATM 4 . In general, banknotes in the ATM 4 are collected in a cassette 41, and banknotes from branches are collected in a transport bag. When bills are collected, a receipt 42 printed with the total amount of collected bills, branch name, ATM number, collection date, etc. is attached. The operator 22 of the sorter 20 sees the receipt 42 and inputs the collection date C106, collection branch name C107, ATM number C108, and collection amount C109 from the user interface device 202 of the sorter 20 before starting the transaction. Information input from the user interface device 202 is recorded in the storage device within the sorter 20 .

As described above, the sorter 20 is a device that sorts banknotes according to their denominations and banknote conditions, and totals the amounts. The state of bills includes fit bills, unfit bills, and counterfeit bills. A fit note is a note that is not very dirty and can be reused. Unfit notes are banknotes that are dirty and cannot be reused. A counterfeit note is a banknote that has been counterfeited or altered.

Items C111 to C119 indicate the respective number of sheets as a result of sorting by the sorter 20 based on the denomination and state of the bills. That is, items C111 to C113 are information on the sorting of 10,000 yen bills, items C114 to C116 are information on the sorting of 5,000 yen bills, and items C117 to C119 are information on the sorting of 1,000 yen bills. Items C111, C114 and C117 indicate the number of fit notes, items C112, C115 and C118 indicate the number of unfit notes, and items C113, C116 and C119 indicate the number of counterfeit notes.

The difference amount C120 indicates the difference in the total amount of one transaction, that is, the difference between the sorted amount C110 and the collected amount C109. The value of the difference amount C120 is normally "0", but if the collected bills contain counterfeit bills or if the operator 22 steals the bills, the sorted amount C110 and the collected amount C109 will match. do not do. When a miscalculated amount occurs, the cause is investigated and evidence is collected through operations within the cash center.

The moving image file C121 is a moving image file when any of counterfeit bill, fraudulent act, or amount mismatch is detected. 10 minutes before and after the time at which any counterfeit note, fraudulent act, or price discrepancy was detected in the video file captured by the surveillance camera 21, and a total of 20 minutes of the situation of the operator 22 is recorded. is a file. The moving image file C121 is an example of "image data of an operator corresponding to a predetermined event".

The number of counterfeit bills C122 is the number of counterfeit bills detected by the sorter 20 during one transaction. The counterfeit note image file C123 is an image file of the detected counterfeit note.

The sorter 20 records the data of items C101 to C123 in a storage area inside the sorter 20, and transmits the data to the management server 1 together with the data of items C206 to C210 described in the serial number list T20 after the transaction is completed.

When receiving the data for one transaction from the sorter 20, the management server 1 generates a transaction ID by numbering the data for one day, associates the generated transaction ID with the data for one transaction, and Record it in the management database T10. When the administrator clicks the transaction ID of the transaction management database T10 on the screen of the administrator terminal 5, the screen transitions to the screen of the serial number list T20.

FIG. 4 shows a serial number list T20 contained in the database 11. As shown in FIG. The serial number list T<b>20 indicates a list of banknote serial numbers recorded in the management server 1 . Items C202 to C210 are information recorded in the storage area of the sorter 20 when banknotes are sorted by the sorter 20, and are transmitted from the sorter 20 to the management server 1 when the transaction is completed.

Comparing the list T20 of FIG. 4 with the database T10 of FIG. 3, transaction ID C201 is the same as transaction ID C101. The operator name C202 is the same as the operator C104. Sorter number C203 is the same as sorter number C105. The sorted amount C204 is the same as the sorted amount C110. The difference amount C205 is the same as the difference amount C120. The date C206 is the same as the date included in the trading start time C102.

Time C207 indicates the time when the banknotes were sorted by the sorter 20 . The serial number C208 indicates a number unique to the banknote written on the face of the banknote. The sorter 20 reads the banknote as an image and records the unique serial number C208 in the storage area within the sorter 20 by recognizing the characters on the image.

The denomination C209 indicates the denomination type of the banknote. The type C210 indicates the state of the banknote. Items C207 to C210 are sent together when the sorter 20 sends data to the management server 1, as described above. When receiving the data from the sorter 20, the management server 1 records the values of the items C207 to C210 in the serial number list T20. Then, the management server 1 associates the transaction management database T10 and the serial number list T20 with the transaction ID.

FIG. 5 is an example of the database T30 that manages fraudulent behavior pattern videos recorded in the management server 1. As shown in FIG. When the surveillance system 3 is introduced, a video file C302 recording various fraudulent actions is created in advance and registered in the database 400. - 特許庁The management server 1 assigns a pattern number C301 and stores it in association with the moving image file C402 so that each fraud pattern can be identified.

FIG. 6 is an example of the database T40 that stores and manages moving image files captured by each monitoring camera 21. As shown in FIG.

The management server 1 assigns a serial number C401 and associates it with the moving image file C405 in order to identify the moving image file captured by each surveillance camera 21 . The surveillance camera number C402 is a number that identifies each surveillance camera 21 . The date C403 is the date when the moving image file was shot. The recording start time C404 is the time when recording was started. The moving image file C405 is the name (storage destination) of the captured moving image file.

FIG. 7 is an example of a database T50 for managing employees, which is included in the database 11 of the management server 1. As shown in FIG.

An employee number C501, which is a number for identifying an employee, is written on the bib 221 worn by the employee (operator 22 and manager 51) and on the hat. The monitoring camera number C502 is a number identifying the monitoring camera 21 assigned to monitor the employee. The name C503 is the name of the employee (only the last name is displayed in FIG. 7). The occupation C504 includes "operator" and "administrator". The face image C505 displays the face of the employee. The email address C506 is the employee's contact email address.

FIG. 8 is a flow chart showing a process (S10) for registering a fraud pattern in the monitoring system 3 (management server 1). By registering in advance the behavior of the operator 22, such as stealing bills, as a fraudulent behavior pattern in the monitoring system 3, the suspicious behavior of the operator 22 can be automatically detected.

If a fraud pattern is to be newly registered (S11: YES), the movie file of the fraud pattern is newly registered in the database T30 shown in FIG. 5 (S12). If the moving image file of the fraud pattern that has already been registered is to be updated (S11: No, S13: YES), the moving image file of the targeted fraud pattern is replaced with a new moving image file, and the database T30 is updated (S14).

If the video file of the registered fraud pattern is to be deleted (S11: NO, S13: NO, S15: YES), the target video file of the fraud pattern is deleted from the database T30 (S16). When the operation (registration operation, update operation, deletion operation) on the database T30 is completed, this processing ends normally (S17: YES).

The database T30 can be updated each time a new cheating technique is found and a video file of the new cheating pattern is created. The moving image file registered in the database T30 may be a file shot in advance at the cash center or factory, or a file shot on the spot using the actual sorter 20 of the cash center 2.

FIG. 9 is a flow chart showing the process (S20) of monitoring the actions of the operator 22 working at the cash center 2 and recording them as a moving image file.

When the preset photographing start time arrives (S21: YES), each monitoring camera 21 starts photographing the target area (the area where the operator 22 works during the sorting work) near the sorter 20 (S22). When the shooting end time arrives (S23: YES), the moving image file is saved and the database T40 shown in FIG. 6 is updated (S24).

If today's work at the cash center 2 is completed (S25: YES), this process ends normally. If the working hours (business hours) of the cash center 2 have not ended (S25: NO), the process returns to step S21, and waits until the next predetermined time zone for photographing arrives.

Predetermined time periods during which the surveillance cameras 21 are photographed include, for example, the time period from 9:00 am to 12:00, the time period from 12:00 pm to 18:00 pm, and the time period from 18:00 to 24:00. can give In other words, in this embodiment, the operating hours of the cash center 2 are divided into a plurality of time zones, and a moving image file of monitoring the operator 22 is created for each time zone and stored in the management server 1 . By creating a moving image file for each time zone, it is possible to quickly search the moving image file for image data that can serve as evidence when an illegal act or the like is detected.

FIG. 10 shows a process (S30) of creating predetermined data as proof of fraudulent activity by the operator 22 by analyzing a video file captured by the surveillance camera 21 and notifying the terminal 5 of the administrator 51 of the data. It is a flow chart. In other words, this process automatically outputs who was doing what when as evidence in the event of a fraudulent act.

The fraud detection unit 10 of the management server 1 checks whether there is a new moving image file, that is, whether there is an unanalyzed moving image file (S31). If there is an unanalyzed moving image file (S31: YES), the fraud detection unit 10 detects that the unanalyzed moving image file contains image data corresponding to any of the fraudulent patterns registered in the database T30. is checked (S32).

If the image data corresponding to the fraud pattern is not detected (S33: NO), the process returns to step S31. If the image data corresponding to the fraud pattern is detected (S33: YES), the fraud detection unit 10 analyzes the image data to identify the operator 22 who performed the action corresponding to the fraud pattern. (S34).

For example, the face of the operator 22 shown in the image data is compared with each face image C505 registered in the employee management database T50, or the ID image displayed on the bib 221 of the operator 22 or the hat is matched. The employee number C501 registered in the employee management database T50 is collated. By these collation processes, it is possible to identify which operator 22 among the employees registered in the employee management database T50 has performed an operation considered to be a fraudulent act.

The fraud detection unit 10 detects the transaction ID C101 from the transaction management database T10 using the operator name/ID C104 and the fraud occurrence time as keys (S35). The fraud detection unit 10 cuts out image data for a predetermined period of time (for example, 10 minutes before and after the time when the fraud occurred, for a total of 20 minutes) based on the time when the fraud occurred from the video file being monitored. , and save it in item C121 in FIG. 3 (S37).

A file name is generated, for example, "F_date_transaction ID.mp4". "F" stands for fraud. "Date" is the date on which the fraudulent act occurred. "Transaction ID" is an ID that identifies a transaction in which fraudulent activity occurred.

The fraud detection unit 10 saves image data (moving image file) extracted as evidence of fraud in the column of the transaction ID C101 in which the fraud occurred among the transaction IDs registered in the transaction management database T10. (S36).

The fraud detection unit 10 notifies the administrator 51 of the occurrence of fraud (S37). For example, an e-mail containing a message (or mail title) indicating that a fraudulent act has occurred and transaction ID C101 in which the fraudulent act occurred is sent to the administrator's 51 e-mail address.

When the administrator terminal 5 receives the e-mail from the management server 1, the administrator 51 accesses the management server 1 using the administrator terminal 5, and uses the transaction ID described in the e-mail as a key to create the data in the database T10. Search for As a result, the administrator 51 can confirm the moving image file C121, the counterfeit image file C123, and other items C102 to C120, C1223, and C201 to C210 at the time of occurrence of counterfeit/fraudulent activity. The administrator 51 can copy evidence data as "predetermined data" to the external memory 6, display it on the display of the user interface unit 16, or print it out from the printer of the user interface unit 16.

Image data as evidence may be reproducibly embedded in the e-mail notifying of the occurrence of fraud, or a link to the image file may be set. As described with reference to FIG. 10, by notifying the fraudulent transaction ID by e-mail, the fraudulent sorting result data and the image data may be associated with each other.

FIG. 11 is a flow chart showing the sorting process (S40) of the sorter 20. As shown in FIG. The operator 22 logs in from the screen of the user interface device 202 of the sorter 20 and inputs predetermined information (S41).

In step S41, the sorter 20 records the name/ID of the operator 22. FIG. The operator 22 inputs a collection date C106, a collection branch name C107, an ATM number C108, and a collection amount C109 into the user interface device 202 as information (predetermined information) about banknotes to be sorted by the sorter 20 . These pieces of input information are stored in the storage area within the sorter 20 . Similarly, the information detected by the sorter 20 is temporarily stored in the storage area within the sorter 20 and then transmitted to the management server 1 .

When the operator 22 presses the transaction start button (S42: YES), the sorter 20 records the transaction start time C102 (S43). When the sorter 20 confirms that banknotes have been set by the operator 22 (S44: YES), the sorter 20 identifies the banknotes, sorts them, and displays the sorted amount C110 on the screen of the user interface device 202 (S45).

The sorter 20 records the sorting amount C110 and the following identification results (S46). That is, the identification results include the number of (10,000 yen bills) fit notes C111, the number (10,000 yen bills) unfit notes C112, the number of (10,000 yen notes) counterfeit notes C113, the number of (5,000 yen notes) correct notes C114, and the number of (5 1,000 yen bill) unfit note C115, number (5,000 yen bill) counterfeit note C116, number (1,000 yen bill) fit note C117, number (1,000 yen bill) unfit note C118, number (1,000 yen bill) counterfeit note C119, time There are C207, serial number C208, denomination C209, and type C210.

When the sorter 20 detects counterfeit bills, the number of counterfeit bills C122 and the counterfeit bill image file C123 are recorded (S46).

When the operator 22 presses the transaction end button (S47: YES), the sorter 20 records the transaction end time C103 (S48). Further, the sorter 20 associates the sorter number C105, the information corresponding to the items C102 to C119, and the information corresponding to the items C207 to C210, C122 and C123, and transmits them to the management server 1 (S48).

FIG. 12 is a flow chart showing the sorting result registration process (S50) executed by the management server 1. As shown in FIG. When the management server 1 receives data from the sorter 20 (S51), it numbers the transaction ID C101 or transaction ID C201 (S52).

The management server 1 updates the transaction management database T10 and the serial number list T20 using the transaction ID C101 or the transaction ID C201 as a key (822). The management server 1 determines whether or not the data received from the sorter 20 contains information on counterfeit bills (S54). (S54: NO), the process proceeds to step S55, which will be described later. If a counterfeit note has been detected (S54: YES), the flow proceeds to a counterfeit note evidence output process (S60), which will be described later.

The management server 1 determines whether or not the amount described on the receipt 42 and the amounts sorted by the sorter 20 do not match (S55). to the evidence output process (S70). If the amount of money does not match (S55: NO), this process ends normally.

FIG. 13 is a flow chart showing the counterfeit evidence output process (S60). The management server 1 detects the surveillance camera name C502 from the database T50 using the operator name/ID C503 that sorted the counterfeit bill as a key (S61). The management server 1 searches the database T40 for the video file C405 that is expected to contain the image data when the counterfeit note was discovered, using the counterfeit note generation date C206, time C207, and surveillance camera name C502 as keys (S62).

The management server 1 cuts out the image data for 10 minutes before and after the time C207 of the counterfeit note generation date C206 (20 minutes in total) from the moving image file C405 and assigns the name C121 (S63). A file name such as "G_date_transaction ID.mp4" can be given to the image data that serves as evidence for finding a counterfeit bill. "G" stands for counterfeit bill. "Date" is the date when the counterfeit bill occurred. "Transaction ID" is the ID of the transaction in which the counterfeit bill occurred.

The management server 1 writes the image data (counterfeit bill video file C121) extracted in step S63 in the column of the transaction ID C101 in which counterfeit bills have occurred among the transaction IDs of the transaction management database T10, and updates the database T10. (S64). Finally, the management server 1 creates an e-mail indicating that a counterfeit bill has occurred, and sends it to the e-mail address C506 of the administrator 51 (S65).

FIG. 14 is a flow chart showing evidence output processing (S70) when an amount mismatch occurs. The management server 1 identifies the monitoring camera 21 corresponding to the sorter 20 in which the price mismatch has occurred (S71), and identifies the moving image file in which the price mismatch has occurred from among the moving image files shot by the monitoring camera 21 (S71). S72).

The management server 1 extracts and saves image data for a predetermined period of time based on the occurrence time of the price discrepancy from the specified moving image file (S73). After updating the transaction management database T10 (S74), the management server 1 notifies the manager 51 by e-mail of the occurrence of the amount mismatch (S75).

FIG. 15 is a flow chart showing the process of printing evidence or storing it in an external memory (S80).

The manager 51 searches the transaction management database T10 or the serial number list T20 using the transaction ID C101 described in the e-mail notifying of the occurrence of fraudulent activity or counterfeit as a key to specify evidence data. (S81).

When printing is instructed (S82: YES), the management server 1 instructs the printer to print the proof data specified in step S82 (S83). A printer directly connected to the management server 1 or a printer connected to the management server 1 via a network may be used. Furthermore, instead of printing on a paper medium, it may be printed as an electronic medium.

When the administrator 51 does not request printing (S82: NO), the management server 1 determines whether copying to the external memory 6 is requested (S84). If the copy to the external memory 6 is requested (S84: YES), the evidence data is written from the memory interface unit 17 to the external memory 6 (S85).

According to this embodiment configured as described above, since the cash center 2 is monitored by the monitoring camera 21 and the captured video file is analyzed in real time, fraudulent acts by the operator 22 can be automatically and quickly discovered. and the reliability of the cash center 2 is improved.

Furthermore, in this embodiment, not only fraudulent acts but also incidents such as discrepancies in monetary amounts and the occurrence of counterfeit bills can be dealt with, and the administrator can easily check the evidence data, thereby supporting the administrator's report creation work. and the usability for the administrator is improved.

In addition, this invention is not limited to embodiment mentioned above. Those skilled in the art can make various additions, modifications, etc. within the scope of the present invention. Features described in the embodiments can be combined as appropriate.

From the description of this embodiment, the sorter 20 can also be expressed as follows.
An automatic cash sorting machine 20 for sorting cash in a cash center 2, which manages sorting result data as a result of sorting work, and image data and serial number of the counterfeit bill when a counterfeit bill is detected during the sorting work. send to device 1,
Automatic cash sorting machine.

From the description of this embodiment, the management server 1 can also be expressed as follows.
A management device 1 that manages a cash center that sorts cash,
receiving a video file captured by a surveillance camera 21 that captures an operator 22 operating an automatic cash sorting machine 20;
receiving sorting result data that is the result of sorting work in the automatic cash sorting machine;
When a predetermined event set in advance is detected by analyzing the moving image file and the sorting result data, the image data of the operator 22 corresponding to the predetermined event is extracted from the moving image file. and a management device for associating the sorting result data corresponding to the predetermined event and the extracted image data among the sorting result data and providing the extracted image data to a predetermined output destination.

1: management server, 2: cash center, 3: monitoring system, 4: ATM, 5: administrator terminal, 6: external memory, 10: fraud detection unit, 11: database, 12: communication unit, 13: sorting result data Acquisition unit 14: video data acquisition unit 15: output unit 16: user interface unit 17: memory interface unit 20: sorter 21: surveillance camera

Claims (5)

A monitoring system for monitoring a cash center that sorts cash,
a surveillance camera that captures an operator who operates the automatic cash sorting machine;
A management device communicatively connected to the automatic cash sorting machine and the surveillance camera, wherein when a predetermined event is detected, the extracting image data of an operator corresponding to a predetermined event, and associating the sorting result data corresponding to the predetermined event among the sorting result data from the automatic cash sorting machine with the extracted image data to obtain a predetermined a management device for providing to an output destination;
with
The predetermined event includes a case where the data indicating the type and amount of cash input to the sorting machine do not match the sorting result of the sorting machine.
Cash center monitoring system. The image data is extracted from the moving image file based on the occurrence time of the predetermined event,
The cash center monitoring system according to claim 1. the predetermined output destination is an information terminal used by an administrator of the cash center;
3. The cash center monitoring system according to claim 2. The surveillance camera shoots a predetermined target area set in advance for a predetermined time period set in advance, and saves the moving image file.
The cash center monitoring system according to any one of claims 1 to 3. A monitoring method for using a computer to monitor a cash center that sorts cash, comprising:
In the cash center, an operator operates a sorting machine,
The computer is communicably connected to a surveillance camera for photographing the operator and the sorting machine,
When the set predetermined event is detected, the computer extracts image data of the operator corresponding to the predetermined event from the video file captured by the surveillance camera, and extracts the image data of the operator corresponding to the predetermined event from the sorting machine. of the sorting result data, the sorting result data corresponding to the predetermined event and the extracted image data are associated with each other and provided to a predetermined output destination ;
The predetermined event includes a case where the data indicating the type and amount of cash input to the sorting machine do not match the sorting result of the sorting machine.
How to monitor cash centers.

Images