JP7448701B1 - Person detection device, person detection method, and person detection program - Google Patents

Abstract

An object of the present invention is to avoid erroneously detecting the presence of a person when an object other than a person is present near an automatic transaction device.
SOLUTION: A person detection device 20 includes an object detection unit 21 that detects the presence of an object within an object detection distance 210 from an automatic transaction device 2, and a captured image 220 that captures the vicinity of the automatic transaction device 2. an acquisition unit 22 that acquires images; a determination unit 23 that determines whether or not there is a person within a predetermined range from the automatic transaction device 2 by analyzing the captured image 220; a setting unit 24 that sets the object detection distance 210 to a value shorter than the current value when the presence of the object is detected and the automatic transaction device 2 determines that no person exists within a predetermined range; Equipped with
[Selection diagram] Figure 4

Description

The present invention relates to a person detection device, a person detection method, and a person detection program.

A large number of automatic transaction devices are used, such as automated teller machines (ATMs) at banks and automatic ticket vending machines at railway stations, which automatically perform transactions such as depositing and disbursing cash and purchasing tickets through user operations. . Technologies have been applied to such automatic transaction devices to solve various problems such as realizing power savings and improving user safety and convenience.

In relation to the above-mentioned technology, Patent Document 1 discloses an automatic transaction device that accepts user operations to conduct transactions, and is capable of changing the detection distance of a customer detection sensor provided on the front of the device. is disclosed. This device detects the customer based on the time from when the customer is detected by the customer detection sensor until the transaction is started, or whether the transaction is started within a predetermined time after the customer is detected by the customer detection sensor. Automatically adjust the detection distance of the detection sensor.

Furthermore, Patent Document 2 discloses an automatic transaction device that includes a handset that outputs voice guidance to a user and receives input from the user. This device includes a medium ejection section that ejects the medium, and controls the medium ejection section to perform ejection control to eject the medium when a predetermined input is received from the user via the handset.

Further, Patent Document 3 discloses a monitoring system for an automatic transaction device in which images can be taken of the vicinity of the operation unit including the operation screen of the automatic transaction device and the customer's operation itself using a surveillance camera. This system controls a surveillance camera so as not to capture images that can identify customers' important confidential information, or performs image processing so that the images become indistinguishable.

Further, Patent Document 4 discloses an interpersonal automatic transaction device that determines whether there is a person waiting to use the device.If it is determined that there is a person waiting to use the device, this device changes the handling operation. By restricting the number of devices, it is possible to avoid problems such as people waiting in line having to line up at other devices.

Japanese Patent Application Publication No. 2011-150554 Japanese Patent Application Publication No. 2014-149708 Japanese Patent Application Publication No. 2008-027167 Japanese Patent Application Publication No. 2003-316961

In order to conserve power, the automatic transaction device described above is placed in a standby state with the display screen turned off when not in use, and when it detects a person approaching to use the device, it automatically detects that person. In some cases, a standby function is provided to cancel the standby state so that the device can be used immediately. Such an automatic transaction device is usually equipped with an object detection sensor for detecting a person in the vicinity using ultrasonic waves or the like.

The object detection sensor detects objects other than people (obstacles) as long as they reflect ultrasonic waves or the like. Therefore, if there is an obstacle within the object detection range of the object detection sensor, or if there is an obstacle that reflects ultrasonic waves etc. even outside the object detection range, the automatic transaction device will actually will continue to falsely detect the presence of a person even though there are no people nearby. In this case, the above-mentioned standby function of the automated transaction device will virtually cease to operate, and maintenance personnel will normally be required to conduct on-site inspections and move obstacles, which will increase maintenance costs for the automated transaction device. become. Therefore, when an object other than a person exists near an automatic transaction device, it is a problem to avoid erroneously detecting the presence of a person. The above-mentioned Patent Documents 1 to 4 cannot solve such problems.

The main purpose of the present invention is to avoid erroneously detecting the presence of a person when an object other than a person exists near an automatic transaction device.

A person detection device according to one aspect of the present invention includes an object detection unit that detects the presence of an object within an object detection distance from an automatic transaction device, and acquires a captured image of the vicinity of the automatic transaction device. an acquisition means; a determination means for determining whether a person is present within a predetermined range from the automatic transaction device by analyzing the captured image; and an object is present within the object detection distance from the automatic transaction device. and setting means for setting the object detection distance to a value shorter than the current value when it is determined that there is no person within a predetermined range from the automatic transaction device.

In another aspect of achieving the above object, a person detection method according to one aspect of the present invention detects, by an information processing device, the presence of an object within an object detection distance from an automatic transaction device, and By acquiring a captured image of the vicinity of the device and analyzing the captured image, it is determined whether there is a person within a predetermined range from the automatic transaction device, and the object detection distance from the automatic transaction device is determined. When it is detected that an object exists within a predetermined range from the automatic transaction device and it is determined that no person exists within a predetermined range from the automatic transaction device, the object detection distance is set to a value shorter than the current value.

In addition, from a further viewpoint of achieving the above object, a person detection program according to one aspect of the present invention includes an object detection process for detecting the presence of an object within an object detection distance from an automatic transaction device, and an acquisition process for acquiring a captured image of the vicinity of the transaction device; a determination process for determining the presence or absence of a person within a predetermined range from the automatic transaction device by analyzing the captured image; If the presence of an object is detected within the range of the object detection distance and the automatic transaction device determines that no person exists within the predetermined range, the object detection distance is set to a value shorter than the current value. The automatic transaction device is caused to perform a setting process for setting the transaction.

Furthermore, the present invention can also be realized by a computer-readable, non-volatile recording medium in which such a person detection program (computer program) is stored.

The present invention can avoid erroneously detecting the presence of a person when there is an object other than a person near the automatic transaction device.

FIG. 1 is a block diagram showing the configuration of a person detection device 10 according to a first embodiment of the present invention. FIG. 6 is a diagram illustrating that the setting unit 14 according to the first embodiment of the present invention changes the value of an object detection distance 183. It is a flowchart (1/2) showing the operation of the person detection device 10 according to the first embodiment of the present invention. It is a flowchart (2/2) showing the operation of the person detection device 10 according to the first embodiment of the present invention. FIG. 2 is a block diagram showing the configuration of a person detection device 20 according to a second embodiment of the present invention. It is a flow chart showing operation of person detection device 20 concerning a 2nd embodiment of the present invention. FIG. 2 is a block diagram showing the configuration of an information processing device 900 that can implement a person detection device according to each embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

<First embodiment>
FIG. 1 is a block diagram showing the configuration of a person detection device 10 included in an automatic transaction device 1 according to a first embodiment of the present invention. The automatic transaction device 1 is used, for example, in an automatic teller machine at a bank, an automatic ticket vending machine at a railway station, or a vending machine installed in various places, for transactions including depositing and paying cash or purchasing tickets and goods. This is a device that automatically performs the following operations according to the user's operations. In order to save power, the automatic transaction device 1 is kept in a standby state with the display screen turned off when not in use, and when it detects a person approaching to use the automatic transaction device 1, the automatic transaction device 1 is turned off. The device has a function to cancel the standby state so that the device can be used immediately.

The person detection device 10 includes an object detection section 11, an acquisition section 12, a determination section 13, a setting section 14, an operation detection section 15, a specification section 16, an output section 17, a storage section 18, an object detection sensor 110, and a camera 120. The object detection unit 11, the acquisition unit 12, the determination unit 13, the setting unit 14, the operation detection unit 15, the identification unit 16, the output unit 17, and the camera 120 are, in order, object detection means, acquisition means, determination means, setting means, and operation. This is an example of a detection means, a identification means, an output means, and an imaging means.

The storage unit 18 is, for example, a storage device such as a RAM (Random Access Memory) or a hard disk 904, which will be described later with reference to FIG. The storage unit 18 stores an object detection result 181, a captured image 182, an object detection distance 183, and a reference time 184. These pieces of information stored in the storage unit 18 will be described later.

The object detection unit 11 receives a signal from the object detection sensor 110 indicating that an object present in the vicinity of the automatic transaction device 1 has been detected. The object detection sensor 110 detects an object by emitting an ultrasonic wave or the like and detecting a reflected wave of the emitted ultrasonic wave reflected by the object. In this embodiment, objects include people and objects other than people (obstacles).

The object detection sensor 110 can change the sensitivity for detecting an object, and detects the presence of an object within the range indicated by the value of the object detection distance 183, which is stored in the storage unit 18 and indicates the sensitivity. That is, the object detection sensor 110 can detect an object that is located further away from the automatic transaction device 1 as the value of the object detection distance 183 becomes longer.

The object detection unit 11 detects the presence of an object within the object detection distance 183, as described above, based on the signal received from the object detection sensor 110. The object detection unit 11 stores an object detection result 181 obtained by detecting an object existing in the vicinity of the automatic transaction device 1 in the storage unit 18.

The acquisition unit 12 acquires a captured image 182 of the vicinity of the automatic transaction device 1 captured by the camera 120, and stores the acquired captured image 182 in the storage unit 18.

The camera 120 may be configured to constantly take images of the vicinity of the automatic transaction device 1 when the automatic transaction device 1 is in a usable state (that is, not in a standby state). Alternatively, the camera 120 may be normally not activated and activated upon receiving a request to acquire the captured image 182 from the acquisition unit 12.

The acquisition unit 12 may also acquire the captured image 182 when the duration of detecting the presence of the object represented by the object detection result 181 exceeds the reference time 184. As described later, the person detection device 10 according to the present embodiment acquires a captured image 182 in order to determine whether an object (obstruction) other than a person existing nearby is detected. When an object other than a person exists near the automatic transaction device 1, the object detection unit 11 continues to detect the object for a long time. On the other hand, a person may stay near the automatic transaction device 1 regardless of whether the automatic transaction device 1 is used or not, but the person's stay is generally temporary. Therefore, if the duration of detecting the presence of the object represented by the object detection result 181 exceeds the reference time 184 (that is, the duration becomes long), there is a presence of a person other than a person near the automated transaction device 1. It is considered that an object (obstacle) exists. Therefore, it is better to acquire the captured image 182 only when the duration of detecting the presence of the object represented by the object detection result 181 exceeds the reference time 184, as the process performed by the automatic transaction device 1 is more efficient. Efficient. Note that the reference time 184 may be the general maximum time that a person is expected to stay in the vicinity of the automatic transaction device 1 continuously, and may be, for example, about 10 minutes.

Note that the reference time 184 is not a fixed value and may be changed depending on the crowding situation of people in the vicinity of the automatic transaction device 1. That is, the reference time 184 may be set by the setting unit 14, which will be described later, to be longer as the time period becomes busier with people, as indicated by information representing the crowding situation of people in the vicinity of the automatic transaction device 1 for each time period. This is because, as mentioned above, the reference time 184, which is determined from the general maximum time that a person is expected to stay in the vicinity of the automated transaction device 1, tends to become longer as the area near the automated transaction device 1 becomes more crowded. Because there is.

Further, in this embodiment, the automatic transaction device 1 includes the camera 120, but the camera 120 does not have to be built into the automatic transaction device 1. The camera 120 may be, for example, a surveillance camera installed to be able to image the vicinity of the automatic transaction device 1. However, in that case, it is assumed that the person detection device 10 and the camera 120 are communicably connected.

The determination unit 13 determines the presence or absence of a person within a predetermined range from the automatic transaction device 1 by analyzing the captured image 182 acquired by the acquisition unit 12 using existing image analysis technology. However, the predetermined range is a range where it is assumed that there is a person who wants to use the automatic transaction device 1, and is, for example, a range within several meters from the automatic transaction device 1.

The determination unit 13 can determine whether or not the captured image 182 includes a person's face (that is, whether or not a person is included) using, for example, existing face recognition technology. The determining unit 13 may also determine whether or not a person is included in the captured image 182 by learning a shape representing the whole or part of a person, without being limited to the face of the person. The determining unit 13 determines that a person exists within a predetermined range from the automatic transaction device 1, for example, when the size of the portion representing the person included in the captured image 182 is larger than the standard.

The setting unit 14 determines that the object detection unit 11 detects that an object exists within the object detection distance 183, and the determination unit 13 determines that there is no person within a predetermined range from the automatic transaction device 1. In this case, the object detection distance 183 is set to a value shorter than the current value. At this time, for example, a plurality of setting values that can be set as the object detection distance 183 may be prepared, and the setting unit 14 may select one of the plurality of setting values.

FIG. 2 is a diagram illustrating how the setting unit 14 according to this embodiment changes the value of the object detection distance 183. In the example shown in FIG. 2, five setting values L1 to L5 (provided that L1<L2<L3<L4<L5) are prepared as a plurality of setting values that can be set as the object detection distance 183. When the value of the object detection distance 183 is the set value L1, the object detection sensor 110 detects an object that is within the distance L1 from the automatic transaction device 1. Similarly, when the value of the object detection distance 183 is the set value L5, the object detection sensor 110 detects an object that is within a distance of L5 from the automatic transaction device 1. That is, the object detection sensor 110 has the lowest sensitivity for detecting an object when the setting value L1 is set as the object detection distance 183, and the object detection sensitivity is the lowest when the setting value L5 is set as the object detection distance 183. The sensitivity is the highest.

For example, as illustrated in FIG. 2, it is assumed that an object (an obstacle) other than a person exists at a distance of L5 from the automatic transaction device 1. In this case, if L5 is set as the current value of the object detection distance 183, the object detection sensor 110 detects an obstacle. The object detection sensor 110 will continue to detect the obstacle unless the obstacle is moved to a place where the distance from the automatic transaction device 1 is greater than L5. Therefore, in this case, the person detection device 10 cannot detect a person approaching to use the automatic transaction device 1.

In this case, the setting unit 14 changes the value of the object detection distance 183 from the setting value L5 to the setting value L4, which is the next longest value after L5. After the value of the object detection distance 183 is changed to L4, the object detection sensor 110 does not detect an obstacle located at a distance of L5 from the automatic transaction device 1. Therefore, after the value of the object detection distance 183 is changed to L4, the person detection device 10 can detect a person approaching to use the automatic transaction device 1.

The operation detection unit 15 detects a user's operation on the automatic transaction device 1. The user's operations include, for example, an input operation on a touch panel or an input button provided in the automatic transaction device 1, an operation for inserting cash from a cash slot, and the like.

When the operation detection unit 15 detects an operation by the user, it is obvious that a person (user) exists near the automatic transaction device 1 without acquiring and analyzing the captured image 182. Therefore, the acquisition unit 12 does not need to acquire the captured image 182 when the operation detection unit 15 detects an operation on the automatic transaction device 1; The captured image 182 may be acquired when there is no such information.

The identification unit 16 detects that the object detection unit 11 detects that an object exists within the object detection distance 183, and the determination unit 13 determines that there is no person within a predetermined range from the automatic transaction device 1. In this case, an analysis is performed to compare the current captured image 182 and the captured image 182 before the presence of an object within the object detection distance 183 was detected. At this time, the identifying unit 16 removes the person included in the current captured image 182 or the previous captured image 182, and then extracts the obstacle included in the current captured image 182 or the previous captured image 182. Then, the identifying unit 16 identifies an obstacle that is not included in the previous captured image 182 but is included in the current captured image 182 as an obstacle to be avoided from being detected by the object detection unit 11. do. That is, the obstacle identified by the identifying unit 16 is an obstacle that is to be moved to another location by the maintenance personnel of the automatic transaction device 1 so that the person detection device 10 does not detect its presence.

The output unit 17 detects that even after the setting unit 14 sets the object detection distance 183 to the lower limit value, the object detection unit 11 detects that an object exists within the object detection distance 183, and the determination unit 13 If it is determined that there is no person within a predetermined range of the automatic transaction device 1, information instructing to conduct an on-site investigation regarding the automatic transaction device 1 is output to the outside. This case corresponds to the case where the distance between the automatic transaction device 1 and the obstacle is shorter than L1 in the example shown in FIG.

In this case, the output unit 17 transmits information instructing to conduct an on-site investigation to, for example, a terminal of a maintenance worker of the automatic transaction device 1. The obstacle is moved to a location farther from the automatic transaction device 1 than L5 by a maintenance worker who is instructed to conduct a field survey, and the automatic transaction device 1 is no longer detected by the obstacle. . The output unit 17 may also transmit the identification result by the identification unit 16 described above to the terminal to the maintenance personnel.

Next, the operation (processing) of the person detection device 10 according to this embodiment will be described in detail with reference to the flowcharts of FIGS. 3A and 3B.

The operation detection unit 15 checks whether the automatic transaction device 1 is being operated by the user (step S101). If the automatic transaction device 1 is in a state where an operation is being performed (Yes in step S102), the process returns to step S101. If the automatic transaction device 1 is not being operated (No in step S102), the acquisition unit 12 checks whether the object detection unit 11 detects that an object exists within the object detection distance 183. (Step S103).

If the object detection unit 11 does not detect that an object exists within the object detection distance 183 (No in step S104), the process returns to step S101. When the object detection unit 11 detects that an object exists within the object detection distance 183 (Yes in step S104), the acquisition unit 12 determines whether the duration of detecting the presence of the object exceeds the reference time 184. Confirm (step S105).

If the duration does not exceed the reference time 184 (No in step S106), the process returns to step S101. If the duration exceeds the reference time 184 (Yes in step S106), the acquisition unit 12 activates the camera 120 and acquires a captured image 182 of the vicinity of the automatic transaction device 1 (step S107).

The determination unit 13 analyzes the captured image 182 acquired by the acquisition unit 12 and determines whether there is a person within a predetermined range from the automatic transaction device 1 (step S109). If a person exists within the predetermined range (Yes in step S109), the process returns to step S101.

If there is no person within the predetermined range (No in step S109), the setting unit 14 checks whether the current value of the object detection distance 183 is the lower limit value (step S110). If the current value of the object detection distance 183 is not the lower limit value (No in step S111), the setting unit 14 sets the object detection distance 183 to a setting value that is one point shorter than the current value (step S112), and the process returns to step S101. Return to

When the current value of the object detection distance 183 is the lower limit value (Yes in step S111), the output unit 17 generates information instructing an on-site investigation regarding the automatic transaction device 1, and transmits the generated information to the maintenance staff's terminal device. (step S113), and the process returns to step S101.

The person detection device 10 according to the present embodiment can avoid erroneously detecting the presence of a person when there is an object other than a person near the automatic transaction device 1. The reason is that when the person detection device 10 detects that an object exists within the range of the object detection distance 183 and the result of analyzing the captured image 182 indicates that there is no person within the predetermined range, This is because the object detection distance 183 is set to a value shorter than the current value.

Below, effects achieved by the person detection device 10 according to the present embodiment will be described in detail.

The object detection sensor included in the automatic transaction device also detects objects other than people (obstacles) as long as they reflect ultrasonic waves or the like. Therefore, if there is an obstacle within the object detection range of the object detection sensor, or if there is an obstacle that reflects ultrasonic waves etc. even outside the object detection range, the automatic transaction device will actually The system continues to falsely detect the presence of a person even though there are no people nearby. In this case, it is usually necessary for maintenance personnel to conduct a field survey and move the obstacles, which increases the maintenance cost of the automatic transaction device. Therefore, when an object other than a person exists near an automatic transaction device, it is a problem to avoid erroneously detecting the presence of a person.

In response to such a problem, the person detection device 10 according to the present embodiment detects the presence of an object within the object detection distance 183 from the automatic transaction device 1. The person detection device 10 acquires a captured image 182 that captures the vicinity of the automatic transaction device 1. The person detection device 10 determines the presence or absence of a person within a predetermined range from the automatic transaction device 1 by analyzing the captured image 182. Then, when the person detection device 10 detects that an object exists within the range of the object detection distance 183 and determines that there is no person within a predetermined range from the automatic transaction device 1, the person detection device 10 detects the object detection distance 183. Set to a value shorter than the current value. That is, when the person detection device 10 continues to detect the presence of an object other than a person, it adjusts the value of the object detection distance 183 so that the object cannot be detected. Thereby, the person detection device 10 can avoid erroneously detecting the presence of a person when there is an object other than a person near the automatic transaction device 1.

Further, the person detection device 10 according to the present embodiment acquires the captured image 182 when the duration of detecting the presence of an object by the object detection unit 11 exceeds the reference time 184. If the duration is within the reference time 184, it is considered that there is a high possibility that a person exists near the automatic transaction device 1. Therefore, if the duration exceeds the reference time 184, the captured image 182 is to be acquired. By narrowing down the number of people, the person detection device 10 can more efficiently perform processing to avoid erroneously detecting the presence of a person.

Further, the person detection device 10 according to the present embodiment acquires the captured image 182 when no operation on the automatic transaction device 1 is detected. When an operation on the automatic transaction device 1 is detected, it is obvious that there is a person nearby, so the person detection device 10 selects the captured image 182 to be acquired when no operation on the automatic transaction device 1 is detected. By doing so, the process of avoiding erroneously detecting the presence of a person can be performed more efficiently.

In addition, the person detection device 10 according to the present embodiment sets the reference time 184 to be longer as the time period becomes more crowded with people, as indicated by the information representing the crowding situation of people in the vicinity of the automated transaction device 1 for each time period. . That is, the person detection device 10 incorrectly detects the presence of a person based on the fact that the time a person stays in the vicinity of the automatic transaction device 1 tends to be longer as the area around the automatic transaction device 1 becomes more crowded. This allows for more appropriate processing to avoid detection.

Further, when the person detection device 10 according to the present embodiment detects that an object exists within the range of the object detection distance 183 and determines that there is no person within a predetermined range from the automatic transaction device 1, , the current captured image 182 and the captured image 182 before detecting the presence of an object within the object detection distance 183 are analyzed. Then, the person detection device 10 identifies objects other than people that are not included in the previous captured image 182 but are included in the current captured image 182. As a result, the person detection device 10 identifies the obstacle that is causing the false detection of the presence of a person, and therefore appropriately supports the maintenance personnel of the automated transaction device 1 in moving the obstacle at the site. be able to.

Further, the person detection device 10 according to the present embodiment detects that an object exists within the range of the object detection distance 183 even after the object detection distance 183 is set to the lower limit value, and the automatic transaction device 1 If it is determined that there is no person within a predetermined range, information instructing an on-site investigation regarding the automatic transaction device 1 is outputted to the outside. Thereby, the person detection device 10 can quickly solve the problem of erroneously detecting the presence of a person, even if it is difficult to avoid the problem by changing the value of the object detection distance 183. It can be solved.

Furthermore, the person detection device 10 according to the present embodiment activates the camera 120 when acquiring the captured image 182 from the camera 120 that captures the captured image 182. That is, the person detection device 10 turns on the power of the camera 120 only when the captured image 182 is needed, so power saving can be achieved.

<Second embodiment>
FIG. 4 is a block diagram showing the configuration of a person detection device 20 according to the second embodiment of the present invention.

The person detection device 20 according to this embodiment includes an object detection section 21, an acquisition section 22, a determination section 23, and a setting section 24. The object detection section 21, the acquisition section 22, the determination section 23, and the setting section 24 are examples of an object detection means, an acquisition means, a determination means, and a setting means, in this order.

The object detection unit 21 detects that an object exists within an object detection distance 210 from the automatic transaction device 2. The object detection distance 210 is, for example, the same information as the object detection distance 183 according to the first embodiment. The object detection section 21 operates in the same manner as the object detection section 11 according to the first embodiment, for example.

The acquisition unit 22 acquires a captured image 220 of the vicinity of the automatic transaction device 2. The captured image 220 is, for example, an image similar to the captured image 182 according to the first embodiment. The acquisition unit 22 operates in the same manner as the acquisition unit 12 according to the first embodiment, for example.

The determining unit 23 determines the presence or absence of a person within a predetermined range from the automatic transaction device 2 by analyzing the captured image 220. The determining unit 13 operates in the same manner as the determining unit 13 according to the first embodiment, for example.

When it is detected that an object exists within the range of the object detection distance 210 and the automatic transaction device 2 determines that there is no person within the predetermined range, the setting unit 24 sets the object detection distance 210 to the current value. Set to a value shorter than the value. The setting unit 24 operates in the same manner as the setting unit 14 according to the first embodiment, for example.

Next, the operation (processing) of the person detection device 20 according to this embodiment will be described in detail with reference to the flowchart in FIG.

The object detection unit 21 detects that an object exists within the object detection distance 210 from the automatic transaction device 2 (step S201). The acquisition unit 22 acquires a captured image 220 of the vicinity of the automatic transaction device 2 (step S202). The determination unit 23 determines the presence or absence of a person within a predetermined range by analyzing the captured image 220 (step S203).

If there is no object within the object detection distance 210 or if a person exists within the predetermined range (No in step S204), the process returns to step S201. If an object exists within the object detection distance 210 and no person exists within the predetermined range (Yes in step S204), the setting unit 24 sets the object detection distance 210 lower than the current value ( Step S205), the process returns to step S201.

The person detection device 20 according to the present embodiment can avoid erroneously detecting the presence of a person when there is an object other than a person near the automatic transaction device 2. The reason is that when the person detection device 20 detects that an object exists within the range of the object detection distance 210 and the result of analyzing the captured image 220 indicates that no person exists within the predetermined range. This is because the object detection distance 210 is set to a value shorter than the current value.

<Hardware configuration example>
In each of the embodiments described above, each part of the person detection device shown in FIG. 1 and FIG. 4 can be realized by dedicated HW (HardWare) (electronic circuit). Furthermore, in FIGS. 1 and 4, at least the configuration described below can be regarded as a functional (processing) unit (software module) of a software program including instructions executed by a processor.
・Object detection units 11 and 21,
- Acquisition units 12 and 22,
- Determination units 13 and 23,
- Setting sections 14 and 24,
・Operation detection unit 15,
・Specific part 16,
・Output section 17,
- Storage control function in the storage unit 18.

However, the division of each part shown in these drawings is a configuration for convenience of explanation, and various configurations may be assumed upon implementation. An example of the hardware environment in this case will be described with reference to FIG. 6.

FIG. 6 is a diagram illustrating the configuration of an information processing device 900 (computer) that can implement a person detection device according to each embodiment of the present invention. That is, FIG. 6 shows the configuration of a computer (information processing device) that can realize the person detection apparatus shown in FIGS. 1 and 4, and shows a hardware environment that can realize each function in the embodiments described above. represent.

The information processing device 900 shown in FIG. 6 includes the following components.
・CPU (Central_Processing_Unit) 901,
・ROM (Read_Only_Memory) 902,
・RAM (Random_Access_Memory) 903,
・Hard disk (storage device) 904,
・Communication interface 905,
・Bus 906 (communication line),
- A reader/writer 908 that can read and write data stored in a recording medium 907 such as a CD-ROM (Compact_Disc_Read_Only_Memory);
- Input/output interface 909 for monitors, speakers, keyboards, etc.

That is, the information processing apparatus 900 including the above components is a general computer in which these components are connected via a bus 906. The information processing device 900 may include a plurality of CPUs 901, or may include a CPU 901 configured with multi-cores. The information processing device 900 may also not include some of the configurations described above.

The present invention, which has been described using the above-described embodiment as an example, supplies a computer program capable of realizing the following functions to the information processing apparatus 900 shown in FIG. The functions are the configurations described above in the block diagrams (FIGS. 1 and 4) referred to in the description of the embodiment, or the functions in the flowcharts (FIGS. 3A, 3B, and 5). The present invention is achieved by subsequently reading the computer program into the CPU 901 of the hardware, interpreting it, and executing it. Further, the computer program supplied to the apparatus may be stored in a readable/writable volatile memory (RAM 903) or a nonvolatile storage device such as ROM 902 or hard disk 904.

Furthermore, in the above case, the method for supplying the computer program into the hardware can adopt a currently common procedure. The procedure includes, for example, a method of installing the program into the device via various recording media 907 such as a CD-ROM, a method of downloading from the outside via a communication line such as the Internet, and the like. In such a case, the present invention can be considered to be constituted by the code constituting the computer program or the recording medium 907 in which the code is stored.

The present invention has been described above using the above-described embodiment as an exemplary example. However, the invention is not limited to the embodiments described above. That is, the present invention can apply various aspects that can be understood by those skilled in the art within the scope of the present invention.

1 Automatic transaction device 10 Person detection device 11 Object detection unit 12 Acquisition unit 13 Judgment unit 14 Setting unit 15 Operation detection unit 16 Specification unit 17 Output unit 18 Storage unit 181 Object detection result 182 Captured image 183 Object detection distance 184 Reference time 2 Automatic Transaction device 20 Person detection device 21 Object detection section 210 Object detection distance 22 Acquisition section 220 Captured image 23 Judgment section 24 Setting section 900 Information processing device 901 CPU
902 ROM
903 RAM
904 Hard disk (storage device)
905 Communication interface 906 Bus 907 Recording medium 908 Reader/writer 909 Input/output interface

Claims (10)

object detection means for detecting the presence of an object within an object detection distance from the automatic transaction device;
acquisition means for acquiring a captured image of the vicinity of the automatic transaction device;
determining means for determining the presence or absence of a person within a predetermined range from the automatic transaction device by analyzing the captured image;
If it is detected that an object exists within the object detection distance from the automatic transaction device, and it is determined that no person exists within a predetermined range from the automatic transaction device, the object detection distance is set to the current value. a setting means for setting a value shorter than
A person detection device comprising: The acquisition means acquires the captured image when the duration of the detection of the presence of the object by the object detection means exceeds a reference time.
The person detection device according to claim 1. further comprising operation detection means for detecting an operation on the automatic transaction device,
The acquisition means acquires the captured image when the operation detection means does not detect an operation on the automatic transaction device.
The person detection device according to claim 1 or claim 2. The setting means sets the object detection distance to one of a plurality of setting values.
The person detection device according to claim 1 or claim 2. The setting means sets the reference time to be longer as the time period becomes more crowded with people, as indicated by information representing the state of crowding of people in the vicinity of the automatic transaction device for each time period;
The person detection device according to claim 2. If it is detected that an object exists within the range of the object detection distance, and the automatic transaction device determines that no person exists within the predetermined range, the current captured image and the object detection distance are Identify objects other than people that are not included in the previous captured image but are included in the current captured image by analyzing the captured image before the presence of an object was detected within the range of The method further includes specifying means for determining.
The person detection device according to claim 1 or claim 2. Even after the object detection distance is set to the lower limit value by the setting means, it is detected that an object exists within the object detection distance, and a person exists within a predetermined range from the automatic transaction device. further comprising an output means for outputting to the outside information instructing to conduct an on-site investigation regarding the automatic transaction device when it is determined that the automatic transaction device is not to be inspected;
The person detection device according to claim 1 or claim 2. The acquisition unit activates the imaging unit when acquiring the captured image from the imaging unit that captures the captured image.
The person detection device according to claim 1 or claim 2. By information processing equipment,
Detects the presence of an object within the object detection distance from the automatic transaction device,
Obtaining a captured image of the vicinity of the automatic transaction device,
Determining the presence or absence of a person within a predetermined range from the automatic transaction device by analyzing the captured image,
If it is detected that an object exists within the object detection distance from the automatic transaction device, and it is determined that there is no person within a predetermined range from the automatic transaction device, the object detection distance is set to the current value. set to a value shorter than,
Person detection method. Object detection processing that detects the presence of an object within an object detection distance from the automated transaction device;
an acquisition process of acquiring a captured image of the vicinity of the automated transaction device;
a determination process of determining the presence or absence of a person within a predetermined range from the automatic transaction device by analyzing the captured image;
If the automatic transaction device detects that an object exists within the object detection distance, and the automatic transaction device determines that no person exists within the predetermined range, the object detection distance is changed to the current object detection distance. Setting process to set a value shorter than the value,
A person detection program that allows a computer to run

Images