JP2010079741A - Transaction monitoring device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a transaction monitoring device for detecting a person who is likely to be scammed in front of an automated teller machine, and for warning the person of a transfer fraud. <P>SOLUTION: This transaction monitoring device is provided with: a microphone for collecting sound near an automated teller machine; a voice decision means for analyzing the voice collected by the microphone, and for deciding whether the voice is one way or not; and a warning means for, when the voice decision means decides that the voice is one way, warning the user using the automated teller machine. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

The present invention relates to a transaction monitoring apparatus that prevents a transfer fraud victim from making a transfer from an automatic transaction apparatus in advance. In particular, a person who may be a crime victim is determined by voice, and caution is urged.

“Transfer fraud” is a general term for fraud cases in which false contents are pointed at an elderly person or the like by telephone or sealed letter, and money transfer is requested. In general, it is “Ole scam”, “Fictional scam”, “Refund scam”, etc.

  Conventionally, as a countermeasure against such a transfer fraud, Patent Document 1 stores a photograph of a user's face and a card number used at any time when using an automatic cash machine such as a CD / ATM in a financial institution store. I'm trying to identify the person I pulled out at a later date.

  Japanese Patent Application Laid-Open No. 2004-228561 detects a suspicious person wearing a mask or a helmet from an image taken by a surveillance camera. Alternatively, a financial institution alerts the user by setting a limit on the transfer amount from the automatic transaction apparatus and further calling at the counter whether there is a fear of a transfer fraud.

JP 2002-183799 A Japanese Patent Laid-Open No. 9-91432

In a recent transfer fraud, for example, a victim may be placed in a wheelchair and guided to an automatic transaction apparatus, and an operation of the automatic transaction apparatus may be guided from a mobile phone to cause a transfer without the victim's knowledge. In this case, the user of the automatic transaction apparatus does not look suspicious like a mask or a helmet because of a good person. Since there is no consciousness of being a scam, it is difficult to determine whether or not the victim is a bank transfer victim using only images. Victims often use a mobile phone by fraudsters and try to perform a transfer operation using an automatic transaction apparatus.
Victims often use a mobile phone by fraudsters and try to perform a transfer operation using an automatic transaction apparatus.

  Therefore, an object of the present invention is to detect a person using a mobile phone in front of an automatic transaction apparatus and call attention to the danger of a transfer fraud.

  In order to solve such a problem, the present invention includes a microphone that collects sound in the vicinity of an automatic transaction apparatus, and voice determination means that analyzes the voice collected by the microphone and determines whether or not the voice is a one-way voice. A transaction monitoring device is provided that includes a warning means for calling attention to a user of the automatic transaction device when the voice determination means determines that the voice is one-way voice. Here, the one-way voice means a voice in which only the voice of the operator can be heard without hearing the voice of the other party of the conversation on a mobile phone or the like.

  Preferably, the sound determination means determines that the sound is a one-way sound if the sound section of the collected sound is separated beyond a predetermined time.

  Further, preferably, the voice determination unit determines that the voice section is one-way voice when the voice section is separated a predetermined time and repeated a predetermined number of times.

  Furthermore, it is preferable that a human body detector for detecting that a person is present in an area where the automatic transaction apparatus can be operated is provided, and the sound determination unit performs the determination when the human body is detected by the human body detector.

  According to this configuration, the user of the automatic transaction apparatus can detect that the operation of the automatic transaction apparatus is guided by a third party using a mobile phone, and can alert the user.

  Hereinafter, a system configuration of an automatic transaction system to which the present invention is applied will be described with reference to the drawings.

FIG. 1 is a system configuration diagram showing the overall configuration of an automatic transaction system.
In the figure, the automatic transaction system includes a central processing unit 3 installed in a management center of a financial institution, and an automatic transaction apparatus 2 and a transaction monitoring apparatus 1 installed in a store such as a financial institution or a convenience store. Alternatively, it is connected via a communication network 9 such as the Internet, and processes a monetary transaction performed by a user of the automatic transaction apparatus 2.

  The transaction monitoring apparatus 1 includes a human body detector 4 such as an infrared sensor or an ultrasonic sensor for detecting an operator of the automatic transaction apparatus 2, a camera 5 for photographing the operator of the automatic transaction apparatus 2, and an automatic transaction apparatus 2. A microphone 6 that collects voices uttered by the operator, a speaker 7 for prompting the operator of the automatic transaction apparatus 2 or a store clerk, and a display 8 are connected. Here, the microphone 6 is a directional microphone in which directivity is given to the position where the operator stands.

  The central processing unit 3 is a so-called server computer that stores customer information in which customer account numbers and deposit amounts are stored in a database, a prohibition list in which account numbers forbidden for transactions are listed, and the like. When the transaction request signal is received from the automatic transaction apparatus 2 through the transaction process, the transaction process signal is processed in accordance with the content of the transaction request signal, customer information, the prohibition list, etc., and the transaction result signal is returned to the automatic transaction apparatus 2. When the abnormality detection signal is received from the transaction monitoring apparatus 1 via the, the prohibition list is edited according to the content of the abnormality detection signal.

  The transaction request signal includes information such as transaction type, transaction amount, and transaction account. If the transaction type is transfer, the transaction account includes at least a transfer destination account that is the account number of the transfer destination. If the transaction type is withdrawal, the transaction account includes the withdrawal account that is the account number of the withdrawal source. The central processing unit 3 compares the transfer account or withdrawal account with the prohibition list in the transaction process, and determines that the transaction is stopped when the transfer account or withdrawal account is included in the prohibition list. Generate and return a transaction result signal to represent. If it is not in the prohibition list or if it is a normal processing request, the requested transaction processing is performed and a control signal or the like is transmitted to the automatic transaction apparatus 2.

  The abnormality detection signal includes information such as transaction type, transaction account, and detection information. The central processing unit 3 additionally stores the transfer account or withdrawal account included in the abnormality detection signal in the prohibition list, and displays information such as a transaction type, a transaction account, and detection information on a display unit (not shown). Notify the manager of the management center of the occurrence of an abnormality.

  The automatic transaction apparatus 2 reads the cash card or passbook possessed by the user and writes to the cash card or passbook as necessary. In addition, the communication network includes a deposit / withdrawal unit that acquires and discharges banknotes and coins, buttons, a touch panel display, a speaker, and the like, and has an operation display unit that receives operations by the user and displays information to the user. 9 to communicate with the central processing unit 3.

  Next, the transaction monitoring apparatus 1 will be described with reference to FIG. FIG. 2 is a functional block diagram of the transaction monitoring apparatus 1. The transaction monitoring apparatus 1 includes a signal processing unit 11, an automatic transaction apparatus I / F 12, a sensor I / F 13, a camera I / F 14, a microphone I / F 15, a speaker I / F 16, a display I / F 17, a communication I / F 18, a storage The unit 19 is configured.

  The signal processing unit 11 determines whether or not the person who operates the automatic transaction apparatus 22 is a suspicious person (whether or not it is related to the victim of the transfer fraud) based on the voice information acquired by the microphone 6. Communication control means 111, suspicious person detection means 112, and alerting means 113 are included. Explanation of each of these means will be described later.

  The automatic transaction apparatus I / F 12 is an interface with the automatic transaction apparatus 2 and acquires information such as an account number operated by a person operating the apparatus from the automatic transaction apparatus 2.

  The sensor I / F 13 is an interface that receives a human body detection signal from the human body detector 4, and outputs a human body detection state to the signal processing unit 11 depending on the presence or absence of the human body detection signal.

  The camera I / F 14 is an interface with the camera 5, and sequentially inputs an image of a person who operates the automatic transaction apparatus 2 captured by the camera 5 to the signal processing unit 11.

  The microphone I / F 15 is an interface with the microphone 6, and outputs the sound collected by the microphone 6 in an area where the automatic transaction apparatus 2 can be operated to the signal processing unit 11. Such sound data is used for analysis of voice information in the suspicious person detection means 112.

  The speaker I / F 16 is an interface with the speaker 7, and outputs voice guidance or the like corresponding to the result processed by the signal processing unit 1 to the speaker 7.

  The display I / F 17 is an interface with the display 8, and outputs guidance and the like by screen display according to the result processed by the signal processing unit 11 to the display 8.

  The communication I / F 18 is an interface with the communication network 9, and when the signal processing unit 11 determines that the person operating the automatic transaction apparatus 2 is a suspicious person, information on the account operated by the person. Is output to the central processing unit 3 via the communication network 9. Similarly, the account information designated by the automatic transaction apparatus 2 is transmitted to the central processing unit 3 via the network 4, and communication is performed to inquire whether the account information is included in the prohibition list 31.

  The memory | storage part 19 shall memorize | store various information required for operation | movement of the transaction monitoring apparatus 1, the program etc. which are performed in the signal processing part 11, and contains a report destination, a suspicious person image, and a background image.

  When the signal processing unit 11 determines that the person who operates the automatic transaction apparatus 2 is a suspicious person or when the person tries to operate an account included in the prohibition list, the report destination reports the fact. The information of the destination is stored. For example, it is information that can be used for identifying the telephone number, I / P address, etc. of the central processing unit 3. Moreover, you may memorize | store the same information so that it can report to the security company which is not illustrated.

  For the suspicious person image, when the signal processing unit 11 determines that the person operating the automatic transaction apparatus 2 is a suspicious person, the face image of the person photographed by the camera 5 is stored for a predetermined number. . The person who operates the automatic transaction apparatus 2 may be stored as a moving image for recording a gesture of the person over a predetermined time after it is determined that the person is a suspicious person.

  The background image is an image used when the signal processing unit 11 detects a suspicious person or a person who uses a mobile phone, and is an image in a state where the operator of the automatic transaction apparatus 2 is not shown, Updated as appropriate.

  Next, each means included in the signal processing unit 11 will be described. Each of these means is a module realized by executing a program stored in the storage unit 19.

  The communication control unit 111 controls general communication with the central processing unit 3 via the communication I / F 18 and the communication network 9. For example, the suspicious person detection means 112 described later communicates the detection of a suspicious person or the possibility of fraud damage to the central processing unit 3.

  The suspicious person detecting means 112 detects whether or not the person who operates the automatic transaction apparatus 2 is a suspicious person that satisfies a predetermined condition. Various determinations as to whether or not the person is a suspicious person can be considered, but in the present embodiment, the person who operates the automatic transaction apparatus 2 according to an instruction of someone is assumed.

  With reference to FIG. 3, the process in the suspicious person detection means 112 in the transaction monitoring apparatus 1 is demonstrated. FIG. 3 is a processing flow of the voice determination unit which is the suspicious person detection unit 112.

  Next, the sound collected by the microphone 6 installed in the vicinity of the automatic transaction apparatus 2 is analyzed, the risk of transfer fraud is determined from the contents of the call, and the process of alerting the operator is described with reference to FIG. I will explain.

  This processing flow is started when the human body detector 4 detects the operator of the automatic transaction apparatus 2. When a human body is detected, sound collection from the microphone 6 is started in the vicinity of the automatic transaction apparatus 2 in step S301.

  Next, in step S302, sound data collected by the microphone 6 is sequentially stored in a frame buffer (not shown) in the storage unit 19. In step S303, a speech section is extracted from the sound data stored in the frame buffer and analyzed.

  The voice section is extracted by passing the sound data through a filter of about 300 Hz to 1000 Hz so that the voice of the person near the automatic transaction apparatus 2 can be distinguished from the voice of the person behind the voice data. A section in which sound data exceeding a predetermined sound pressure level continues for about 200 ms to 600 ms is extracted as a voice section.

  Next, in step S304, it is determined whether the time interval between the temporally adjacent speech sections is 5 seconds or longer. The 5 seconds is assumed to be a time when an instruction is received from the other party of the mobile phone. That is, it is a time that is considered to be spoken by the other party who instructs the operator how to operate the personal identification number, bank account, and the like. Therefore, this time is not limited to 5 seconds, and may be a time derived empirically.

  Next, if there is one voice section having a separation time of 5 seconds or more, the process proceeds to step S305. On the other hand, if there is no voice section having a separation time of 5 seconds or more, the process proceeds to step S308.

  In step S305, 1 is added to the counter. Then, it is determined whether or not the counter exceeds 8 (step S306). The counter “8” is the number of times determined that the operator will wait for an instruction using the mobile phone at least eight times until the transfer process is completed. This value is appropriately determined by experiment and experience, and is not limited to 8.

  If the counter does not exceed 8 in step S306, the process proceeds to step 308. On the other hand, if the counter exceeds 8, in step S307, the user is determined to be a suspicious person who is highly likely to be in a bank transfer fraud using a mobile phone.

  At step S307, the alerting means 113 alerts the person with a suspicious person.

  Here, the alerting means 113 will be described. The alerting means 113 is guided to operate the automatic transaction apparatus 2 from the other party using a mobile phone if a predetermined number of voice intervals that are separated by a predetermined time or more are counted in step S306 in step S304. Judge that there is a fear. In this way, it is a means for calling attention to an operator who may be a transfer fraud.

  In other words, if it is determined as “Yes” in step S306, the display unit 17 is caused to display a predetermined guidance to pay attention to the transfer fraud. Furthermore, voice guidance is played from the speaker 7. That is, the display unit displays the fact that there is a possibility of wire fraud on the display unit via the display unit I / F 17 and alerts someone to consult. Then, the counter is reset and the process proceeds to step S308.

  In step S308, it is determined whether or not the human body detector 4 continues human body detection. Here, if the human body detection continues, the process returns to step S302 and the same processing is continued. On the other hand, if the human body detection is not continued, the sound collection is finished and the present process is finished.

  In the present embodiment, the contents of the speech within the speech section are not examined, but the accuracy is further improved by analyzing whether the speech data of the specific section is included in the speech data of the speech section. Here, the specific word is a word uttered by the operator in response to an operation instruction of the automatic transaction apparatus 2 such as “Yes”, “Yes”, “Pressed”, or the like. The specific word is stored in advance in the storage unit 19 as a voice model.

  And between step S303 and step S304, by collating the voice feature of the voice data in the extracted voice section and the voice feature of the specific word stored in the storage unit 19, the specific word is found in the voice section. Analyze whether it is included.

  Then, the number of speech sections including the specific word may be counted by the counter in step S305. In this case, the number corresponding to 3 in step S306 is determined based on the number of times that “Yes”, “Yes”, etc. are likely to be issued when an operation instruction of the automatic transaction apparatus 2 is received by the mobile phone.

  When the present invention is applied to an automatic transaction system, it is possible to prevent a transfer fraud using an automatic transaction system and a mobile phone in advance.

It is the whole automatic transaction system lineblock diagram concerning this embodiment. 2 is a functional block diagram of a transaction monitoring apparatus 1. FIG. It is a processing flow of a voice determination means.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Transaction monitoring apparatus 11 Signal processing part 111 Communication control means 112 Suspicious person detection means 113 Attention means 19 Storage part 2 Automatic transaction apparatus 3 Central processing unit 4 Human body detector 5 Camera 6 Microphone 7 Speaker 8 Display 9 Communication network

Claims (4)

A microphone that collects sound in the vicinity of the automatic transaction device;
A sound determination unit that analyzes the sound collected by the microphone and determines whether the sound is a one-way sound;
An alerting means for alerting a user of the automatic transaction apparatus when the speech determination means determines a one-way speech;
A transaction monitoring device.   The transaction monitoring apparatus according to claim 1, wherein the voice determination unit determines that the voice section of the collected sound is a one-way voice if the voice section of the collected sound is separated beyond a predetermined time.   The transaction monitoring apparatus according to claim 1, wherein the voice determination unit determines that the voice section is a one-way voice when a voice section is separated a predetermined time and repeated a predetermined number of times. Furthermore, a human body detector that detects that there is a person in an area where the automatic transaction apparatus can be operated is provided.
The transaction monitoring apparatus according to claim 1, wherein the voice determination unit performs determination when a human body is detected by the human body detector.

Images