TWI730374B - Automatic teller machine warning system - Google Patents

Abstract

An automatic teller machine warning system comprises an automatic teller machine, an image capturing component, a face feature comparison unit, a processor and a warning device. The image capturing component is disposed in the automated teller machine and continuously captures a plurality of face images. The face feature comparison unit determines whether the face images have the same face feature and outputs a first confirmation signal when the face images have the same face feature. The processor determines whether the total time of the face images is greater than an upper time limit, when the total time is greater than the upper time limit, the processor outputs a second confirmation signal, and the warning device reads the second confirmation signal to output a warning signal.

Description

ATM Early Warning System

The present invention relates to an automatic teller machine, especially an automatic teller machine with an early warning function.

In modern society, the public can not only withdraw money through ATMs, but also provide services such as deposits, payment or transfers, and the number of ATM bases is also increasing, which has become an indispensable part of modern society. The existing ATMs are only equipped with video recording facilities, and the video and audio content of the video is stored separately by each financial institution. With regard to financial card theft cases, rampant fraudulent group riders, and increasingly sophisticated criminal equipment, the existing security mechanisms of ATMs are fundamentally difficult to prevent, causing losses to financial institutions and customers, and serious public security problems.

In view of this, there is indeed a need for an improved ATM system in practice, which can at least solve the above shortcomings.

The present invention provides an automatic teller machine early warning system, whose detection and alarm mechanism for abnormal events can better target current criminal methods and effectively prevent the automatic teller machine from being stolen.

According to an embodiment of the present invention, an automatic teller machine early warning system is provided, which includes an automatic teller machine, an image capturing component, a facial feature comparison unit, a processor, and a warning device. The image capturing component is arranged in the automated teller machine, and the image capturing component continuously captures multiple facial images. The facial feature comparison unit is electrically connected to the image capturing element and determines whether the facial images have the same facial features. When the facial images have the same facial features, the facial feature comparison unit outputs a first confirmation signal. The processor is electrically connected to the facial feature comparison unit, and when the facial feature comparison unit outputs the first confirmation signal, the processor determines whether the total time spanned by the facial images is greater than the upper limit of the time. When the total time is greater than the upper limit of the time, the processor outputs a second confirmation signal. The warning device is arranged in the automatic teller machine and is electrically connected to the processor. When the processor outputs the second confirmation signal, the warning device reads the second confirmation signal to output the warning signal.

The automatic teller machine early warning system provided by the present invention can determine whether the person operating the automatic teller machine is the same person based on the detected facial image. In addition, the ATM further records the login time of the financial card at the ATM to determine whether the login time of the financial card at the ATM is abnormal. Furthermore, the number of different accounts logged into the ATM is recorded to determine whether the number of financial cards logged into the ATM is abnormal. Through multiple judgment mechanisms, the processor’s accuracy in judging whether financial theft has occurred is improved, and the number of misjudgments is effectively reduced, thereby preventing the warning device from issuing warning signals at the wrong time and causing a waste of hardware resources.

The above description of the content of the present invention and the description of the following embodiments are used to demonstrate and explain the spirit and principle of the present invention, and to provide a further explanation of the scope of the patent application of the present invention.

The detailed features and advantages of the present invention will be described in detail in the following embodiments. The content is sufficient to enable anyone familiar with the relevant art to understand the technical content of the present invention and implement it accordingly, and according to the content disclosed in this specification, the scope of patent application and the drawings. Anyone who is familiar with relevant skills can easily understand the purpose and advantages of the present invention. The following examples further illustrate the viewpoints of the present invention in detail, but do not limit the scope of the present invention by any viewpoint.

Fig. 1 is a schematic diagram showing the early warning system of an automatic teller machine according to the first embodiment of the present invention. As shown in Figure 1, the automatic teller machine early warning system includes an automatic teller machine 10 and a bank server host 20. The automatic teller machine 10 includes a reading unit 11, an image capturing element 12, a warning device 13, and a first network connection interface 14. The server host 20 includes a facial feature comparison unit 21, a processor 22 and a second network connection interface 23. The first network connection interface 14 and the second network connection interface 23 are connected to the network, so that the automatic teller machine 10 and the bank server host 20 can communicate in two directions.

The reading unit 11 of the automatic teller machine 10 is electrically connected to the image capturing element 12 and the first network connection interface 14. When the user of the automatic teller machine 10 inserts a financial card, the reading unit 11 reads the financial card account number to output It can signal to the image capturing element 12 to drive the image capturing element 12 to capture images. The image capturing element 12 can be, for example, a hidden camera, and is used to continuously capture images. The image capturing element 12 is arranged at an appropriate height on the ATM 10 so that the shooting angle range of the image capturing element 12 covers the face of the operator of the ATM 10. For example, the image capturing element 12 is arranged on the frame of the display screen of the ATM 10. In order to reduce the blind spots in shooting, the number of image capturing elements 12 can be multiple and they can be arranged at different positions on the ATM 10, for example, a plurality of image capturing elements 12 are respectively arranged on the upper frame, the left frame, and the left frame of the display screen. Right frame.

The facial feature comparison unit 21 of the bank server 20 can be, for example, a Field Programmable Gate Array (FPGA) that stores facial recognition algorithms and is electrically connected to the processor 22 and the second network Connection interface 23. In an embodiment of the facial feature comparing unit 21, when the face image is a two-dimensional image or a three-dimensional image, the facial feature comparing unit 21 obtains at least one face image from each face image from the image capturing element 12 Facial features (such as eyes, ears, mouth, nose, facial contours) in two-dimensional space or three-dimensional space, the shape, position, or the relative position between several facial features. After that, the facial feature comparison unit 21 stores the facial features of each facial image in the storage device of the bank server 20. The facial feature comparison unit 21 compares the facial features of multiple consecutive face images, and if it is determined that the facial features of the facial images are all the same, it verifies that the facial images are all the same person and outputs The first confirmation signal is sent to the processor 22.

The processor 22 may be, for example, a central processing unit (CPU), a digital signal processor (DSP), a microprocessor (MPU), a graphics processing unit (GPU) or a microcontroller (MCU), and the processor 22 is electrically connected to the 2. The network connection interface 23 continuously records the appearance time of each face image. The processor 22 presets an upper limit of time. When the processor 22 receives the first confirmation signal, the processor 22 determines whether the total time that the facial images appear is greater than the upper limit of time. When the processor 22 confirms that the total time of appearance of the facial images is greater than the preset time limit, the processor 22 outputs a second confirmation signal to the second network connection interface 23. In addition, the processor 22 also presets a quantity threshold. When the processor 22 confirms that the total time of the appearance of the facial images is not greater than the preset upper limit of time, the processor 22 further determines whether the number of different accounts logged into the ATM 10 in the time interval of the appearance of the facial images is greater than Quantity threshold. When the processor 22 confirms that the number of different accounts logged into the ATM 10 is greater than the number threshold, the processor 22 also outputs a second confirmation signal to the second network connection interface 23.

The warning device 13 is electrically connected to the first network connection interface 14, and when the warning device 13 receives the second confirmation signal from the processor 22, the warning device 13 outputs a warning signal. The warning signal includes, for example, a sound signal and a light signal, and the light signal has a flicker frequency, and the sound signal may include a string of warning signs, thereby achieving a deterrent effect.

Fig. 2 is a schematic diagram showing an early warning system of an automatic teller machine according to a second embodiment of the present invention. The embodiment of FIG. 2 is mostly the same as the embodiment of FIG. 1, the difference is that the facial feature comparison unit 15 and the processor 16 are directly installed in the ATM 10, eliminating the need for the ATM 10 to transmit data to the bank server 20 via the network. time. Also, because data transmission does not require a network connection, even if the network connection between the ATM 10 and the bank server host 20 fails, the early warning mechanism on the ATM 10 can still operate normally.

FIG. 3 is a flow chart showing the early warning method of the automatic teller machine according to the first embodiment of the present invention. As shown in FIG. 3, in step S10, the reading unit 11 of the automated teller machine 10 determines whether a financial card account number has been read. If the account number of the financial card is read, step S11 is entered. If the financial card account number is not read, the warning method is ended. In step S11, when the reading unit 11 of the ATM 10 reads the financial card account number, the reading unit 11 outputs an enabling signal to the image capturing element 12. It can be seen that only when the financial card is inserted into the ATM 10, the image capturing component 12 can be activated. In step S12, the image capturing component 12 of the automated teller machine 10 is used to continuously capture a plurality of facial images. In step S13, the facial feature comparison unit 21 determines whether the facial images have the same facial features. When the facial feature comparison unit 21 confirms that the facial images have the same facial features, it means If the user of the automated teller machine 10 is the same person, the step S14 is executed continuously. Conversely, when the facial feature comparison unit 21 confirms that the facial features of the facial images are different, it means that the users of the ATM 10 are not the same person, and the warning method ends. In step S14, the facial feature comparison unit 21 outputs a first confirmation signal to the processor 22. In step S15, after the processor 22 receives the first confirmation signal, the processor 22 determines whether the total time of the appearance of the facial images is greater than the upper limit of time preset by the processor 22. When the processor 22 confirms that the total time is greater than the upper limit of time, step S16 is executed. When the processor 22 confirms that the total time is not greater than the upper limit of the time, it means that the same person operates the automated teller machine 10 within the normal time, and the early warning method ends. In step S16, the processor 22 outputs a second confirmation signal to the warning device 13. In step S17, after the warning device 13 of the ATM 10 obtains the second confirmation signal, the warning device 13 outputs a warning signal. In one embodiment, the warning signal output by the warning device 13 will be transmitted to the security center through the first network connection interface 14 so as to call the staff of the security center for processing. In another embodiment, the warning device 13 continuously outputs the warning signal, so that the user of the ATM 10 can hear the warning signal. The warning device 13 stops sending the warning signal until the reading unit 11 cannot read the financial card account number. This means that the warning device 13 will continuously warn the user of the ATM 10 until he stops using the financial card to operate the ATM 10. The above-mentioned security early warning mechanism is to simulate a situation in which an unscrupulous person collects fingerprints staying on the keys of the ATM 10, or a situation in which a stolen financial card is used to transfer cash to multiple accounts.

4A and 4B are flowcharts of the automatic teller machine early warning method according to the second embodiment of the present invention. As shown in FIGS. 4A and 4B, steps S20 to S27 are substantially the same as steps S10 to S17 in FIG. 3, and the embodiment of FIGS. 4A and 4B further includes step S28. When the processor 22 confirms that the total time is not greater than the upper limit of time, step S28 is executed. In step S28, the processor 22 determines whether the number of different accounts logged into the ATM 10 in the total time is greater than a number threshold. When the processor 22 confirms that the number of different accounts is greater than the number threshold, it means that an abnormal number of financial cards continue to log in to the ATM 10, and step S26 is executed. When the processor 22 confirms that the number of different accounts is not greater than the number threshold, it means that the number of financial cards logged in the ATM 10 is within the normal range, and the early warning method is ended. The above-mentioned early warning mechanism is to simulate a situation in which an unscrupulous person uses multiple stolen financial cards to conduct transactions at the ATM 10.

5A and 5B are flowcharts of an automatic teller machine early warning method according to an embodiment of the present invention. As shown in FIGS. 5A and 5B, steps S30 to S38 are substantially the same as steps S20 to S28 in FIGS. 4A and 4B. The embodiment of FIGS. 5A and 5B further includes steps S39 to S42. When the processor 22 confirms that the number of different accounts is not greater than the number threshold, step S39 is executed. In step S39, the bank server host 20 stores a blacklist data containing multiple facial images with financial criminals, and the facial feature comparison unit 21 determines whether the facial images meet the blacklist data Face image inside. If the facial feature comparison unit 21 confirms that the facial images match the facial images in the blacklist data, step S40 is executed. If the facial feature comparison unit 21 confirms that the facial images do not match the facial images in the blacklist data, the early warning method is ended. In step S40, the facial feature comparison unit 21 sends a third confirmation signal to the processor 22. In step S41, after the processor 22 receives the third confirmation signal, the processor 22 sends a fourth confirmation signal to the warning device 13. In step S42, after the warning device 13 reads the fourth confirmation signal, the warning device 13 sends a warning signal. The above-mentioned early warning mechanism is to add a checkpoint to prevent skilled fraud group drivers.

In other embodiments, after the warning device 13 sends out the warning signal, the processor 22 of the bank server host 20 further sends a notification message to the police station host closest to the ATM 10 according to the address of the ATM 10, thereby calling the police Go to the ATM 10 for processing.

In other embodiments, when the number of ATMs 10 is multiple, the processor 14 determines whether the total number of ATMs 10 with the same financial card account in different locations is greater than an upper limit. When the total number of ATMs 10 in different locations is greater than the upper limit of the number, the processor 22 sends a second confirmation signal to the warning device 13 on the ATM 10 where the financial card is currently logged in, so as to drive the warning device 13 to issue a warning signal. The above-mentioned early warning mechanism is to simulate the situation where an unscrupulous person uses a stolen financial card to conduct transactions at multiple ATMs.

The automatic teller machine early warning system provided by the present invention can determine whether the operator of the automatic teller machine is the same person based on the detected facial image. In addition, the ATM further records the total time for the financial card to log in to the ATM, so as to determine whether the login time of the financial card at the ATM is abnormal. Furthermore, the number of different accounts logged into the ATM is recorded to determine whether the number of financial cards logged into the ATM is abnormal. Through the setting of multiple early warning mechanisms, the processor’s correct probability of determining whether financial theft has occurred is increased, and the number of misjudgments is effectively reduced, thereby avoiding the warning device from issuing warning signals at the wrong time and causing a waste of hardware resources.

Although the present invention is disclosed in the foregoing embodiments, it is not intended to limit the present invention. All changes and modifications made without departing from the spirit and scope of the present invention fall within the scope of the patent protection of the present invention. For the scope of protection defined by the present invention, please refer to the attached scope of patent application.

10: ATM 11: Read components 12: Image capture component 13: Warning device 14: The first network connection interface 20: Bank server host 15, 21: Facial feature extraction unit 16, 22: processor 23: The first network connection interface

Fig. 1 is a schematic diagram showing the early warning system of an automatic teller machine according to the first embodiment of the present invention. Fig. 2 is a schematic diagram showing an early warning system of an automatic teller machine according to a second embodiment of the present invention. FIG. 3 is a flow chart showing the early warning method of the automatic teller machine according to the first embodiment of the present invention. 4A and 4B are flowcharts of the automatic teller machine early warning method according to the second embodiment of the present invention. FIG. 5A and FIG. 5B are flowcharts of the automatic teller machine early warning method according to the third embodiment of the present invention.

10: ATM

11: Reading unit

12: Image capture unit

13: Warning device

14: The first network connection interface

20: Bank server host

21: Facial feature comparison unit

22: processor

23: The second network connection interface

Claims (3)

An automatic teller machine early warning system, comprising: an automatic teller machine; an image capturing element arranged in the automatic teller machine; the image capturing element continuously captures a plurality of facial images; a facial feature comparison unit, which is electrically connected The image capturing element determines whether the facial images have the same facial features, and when the facial images have the same facial features, the facial feature comparison unit outputs a first confirmation signal; a process The processor is electrically connected to the facial feature comparison unit. When the facial feature comparison unit outputs the first confirmation signal, the processor determines whether the total time that the facial images appear is greater than a time limit. When the total time is greater than the upper limit of the time, the processor outputs a second confirmation signal; and a warning device is arranged on the ATM and is electrically connected to the processor. When the processor outputs the second confirmation signal, the processor The warning device reads the second confirmation signal to output a warning signal; wherein when the total time is less than or equal to the upper limit of time, the processor further judges the difference in logging in to the ATM in the time interval in which the facial images appear Whether the number of account numbers is greater than a number threshold, when the processor confirms that the number of different account numbers is greater than the number threshold, the processor outputs the second confirmation signal to the warning device; and a reading unit that reads A financial card account outputs an enabling signal to the image capturing element, and the enabling signal drives the image capturing element to capture the facial images. The automatic teller machine early warning system according to claim 1, wherein the warning signal includes a sound signal and a light signal, and the light signal has a flicker frequency. The automatic teller machine early warning system according to claim 1, wherein the facial feature comparison unit and the processor are arranged in a bank server host, and the bank server host communicates with the automatic teller machine through a network.

Images