Security system, security method and program

Abstract

To provide a security system, a security method, and a program that are capable of estimating a possibility that a fraudulent act was performed during a patrol on the basis of biological signals of a patrolling person.SOLUTION: A security system comprises: a first acquisition unit that acquires a biological signal from a measurement device which detects biological signals related to a patrolling person; a calculation unit that calculates a predetermined feature value from the biological signal acquired by the first acquisition unit; an analysis unit that determines at least either one of a consciousness state and an emotion state of the patrolling person by analyzing the feature value calculated by the calculation unit; and an estimation unit that estimates a possibility that a fraudulent act was performed by the patrolling person on the basis of change in at least either of the consciousness state and the emotion state determined by the analysis unit.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a security system, a security method, and a program. [Background technology]

[0002] Conventionally, security systems typically use security terminals installed in individual buildings and other spaces to be guarded. When a security terminal detects an abnormality, such as the presence of a person or the outbreak of a fire, it notifies a security company or other entity of the abnormality. In such security systems, security guards or other patrol personnel patrol the building on behalf of the homeowner or other entity when the homeowner or other entity is absent, such as at night or on holidays, or when an abnormality is reported, to check the situation. However, while patrols are primarily intended to prevent and deter criminal acts, such as intrusions into buildings by suspicious individuals, there is no guarantee that the patrol personnel themselves will not commit fraud, such as theft. Furthermore, since the patrol personnel are authorized to enter the building, the security system installed in the building cannot detect the patrol personnel's fraudulent behavior. Therefore, a system is needed to detect and deter fraudulent behavior by patrol personnel during patrol work.

[0003] For example, Patent Document 1 discloses a technology that contributes to preventing fraudulent activities. Patent Document 1 includes a central device and at least one store device connected to the central device via a communication line. The store control device acquires the date and time of entry into the store, the date and time of exit from the store, and facial information for each person from a store face authentication unit and transmits them to the central device along with store information. The store control device also acquires the detected movement path information for each person and the corresponding person's facial information from a movement path detection unit and transmits them to the central device along with store information. The store control device also transmits information indicating the presence of a suspicious person from a suspicious behavior detection unit, facial information of the person corresponding to the movement path information determined to be a suspicious person, and location information, along with store information, to the central device as suspicious information. Furthermore, upon receiving information about a lost product and information about the date, time, and time of the loss from the product detection device, the store control device transmits this information, along with store information, to the central device as lost information. The central device determines the suspiciousness of each person based on the information transmitted from the store control device. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 2018-173855 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the technology described in Patent Document 1 detects fraudulent activity by analyzing images from surveillance cameras, etc., but there is a problem in that it is difficult to install cameras in all buildings where patrol personnel may be performing patrol work and determine from camera images whether the patrol personnel's work is fraudulent or not.

[0006] The present invention has been made in consideration of the above, and aims to provide a security system, security method, and program that can estimate the possibility of fraudulent activity during patrol work based on the biological signals of patrol personnel. [Means for solving the problem]

[0007] In order to solve the above-mentioned problems and achieve the object, the present invention is characterized by comprising a first acquisition unit that acquires a biological signal related to a patrolman from a measuring device that detects the biological signal; a calculation unit that calculates a predetermined feature from the biological signal acquired by the first acquisition unit; an analysis unit that determines at least one of the patrolman's state of consciousness or emotional state by analyzing the feature calculated by the calculation unit; and an estimation unit that estimates the possibility of the patrolman committing fraudulent acts based on changes in at least one of the state of consciousness or the emotional state determined by the analysis unit. [Effects of the Invention]

[0008] According to the present invention, it is possible to estimate the possibility that fraudulent activity has occurred during patrol based on the biological signals of the patrol person. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram illustrating an example of the overall configuration of a security system according to an embodiment. [Figure 2] Figure 2 illustrates the changes in consciousness and emotion during patrol work. [Figure 3] FIG. 3 is a diagram illustrating an example of a security target space. [Figure 4] Figure 4 shows an example of changes in consciousness and emotion during patrol work when there is no possibility of fraud. [Figure 5] Figure 5 shows an example of changes in consciousness and emotion during patrol work when fraud is detected and cameras are present. [Figure 6]Figure 6 shows an example of changes in consciousness and emotion during patrol work when fraud is committed and no cameras are present. [Figure 7] FIG. 7 shows an example of changes in consciousness and emotion during patrol work when there is no possibility of fraud and appropriate measures have been taken to address the cause of the alert. [Figure 8] Figure 8 shows an example of changes in consciousness and emotion during patrol work when there is no possibility of fraud and an inappropriate response is taken to the cause of the alert. [Figure 9] FIG. 9 is a diagram illustrating an example of a hardware configuration of an information terminal according to an embodiment. [Figure 10] FIG. 10 is a diagram illustrating an example of a hardware configuration of a management server according to an embodiment. [Figure 11] FIG. 11 is a diagram illustrating an example of a functional block configuration of a security system according to an embodiment. [Figure 12] FIG. 12 is a diagram illustrating an example of a distribution of feature amounts based on a biological signal. [Figure 13] FIG. 13 is a diagram for explaining the calculation operation of the amount of change in the level of awareness in the awareness analysis unit. [Figure 14] FIG. 14 illustrates the distribution of emotional states based on the emotional analysis results. [Figure 15] FIG. 15 is a diagram showing an example of a method of emotion analysis in the emotion analysis unit. [Figure 16] FIG. 16 is a diagram showing an example of a method of emotion analysis in the emotion analysis unit. [Figure 17] FIG. 17 is a diagram illustrating an example of operation history information. [Figure 18] FIG. 18 is a flowchart showing an example of the flow of patrol work when the security system according to the embodiment is used. [Figure 19] FIG. 19 is a flowchart showing an example of the flow of fraud estimation and confirmation processing in the security system according to the embodiment. [Figure 20] FIG. 20 is a diagram showing an example in which some functions of a security system according to a modified example are replaced by processing using a learning model based on machine learning. DETAILED DESCRIPTION OF THE INVENTION

[0010] Below, embodiments of a security system, a security method, and a program according to the present invention will be described in detail with reference to the drawings. Furthermore, the present invention is not limited to the following embodiments, and the components in the following embodiments include those that would be easily conceived by a person skilled in the art, those that are substantially the same, and those that are within the scope of what is called equivalents. Furthermore, various omissions, substitutions, modifications, and combinations of the components can be made without departing from the spirit of the following embodiments.

[0011] (Overall configuration of security system) FIG. 1 is a diagram showing an example of the overall configuration of a security system according to an embodiment. FIG. 2 is a diagram illustrating changes in consciousness and emotions during patrol work, with the horizontal axis representing time and eight waves representing electroencephalogram waveforms. FIG. 3 is a diagram showing an example of a space to be guarded. An overview of the overall configuration and operation of security system 1 according to this embodiment will be described with reference to FIGS. 1 to 3.

[0012] The security system 1 shown in FIG. 1 is a system that estimates the possibility of misconduct by a patrolman in the security space STS by analyzing biological signals such as electroencephalograms acquired from the patrolman patrolling the security space STS. As shown in FIG. 1, the security system 1 includes an information terminal 10 carried by the patrolman and a management server 20 at a monitoring center. The patrolman patrols the security space STS while wearing a helmet H equipped with an electroencephalograph (EEG) E. The electroencephalograph E is a measuring device that measures the neural activity state of the brain by detecting weak electrical signals (brain waves) emitted from the human brain using multiple electrodes. Note that the electroencephalograph E is not limited to being built into the helmet H, but may be any device that can be worn or carried in a manner that allows the patrolman's brain waves to be detected. Furthermore, the measuring device for measuring the biosignals of the patrolman is not limited to the electroencephalograph E, but may also be a magnetoencephalogram (MEG) or a brain activity sensor that uses near-infrared light to measure changes in blood flow in brain tissue.

[0013] A security terminal ST equipped with various sensors, such as motion sensors, cameras, image sensors, and door sensors, and a communication device, is installed at various locations in the guarded space STS, enabling detection of the presence of humans and other objects in the guarded space STS. At least one sensor is installed in the guarded space STS. In the example shown in FIG. 3 , the sensors include motion sensors ST-M1, ST-M2, and ST-M3 that detect humans and other objects, and cameras ST-C1 and ST-C2 that capture images to detect humans and other objects. When a sensor detects a human, such as a patrolman, the communication device transmits detection information, including the sensor's identification information and the detection time, to the management server 20 in the monitoring center via the network N. The sensor identification information is, for example, information assigned to each sensor in advance to identify each sensor. In this embodiment, it refers to "ST-M1," "ST-M2," "ST-M3," "ST-C1," and "ST-C2" assigned to the motion sensors and cameras. The human, etc., may be, for example, a patrolman. The network N is a network such as the Internet or a LAN (Local Area Network) conforming to protocols such as TCP (Transmission Control Protocol) / IP (Internet Protocol). Note that the network N may include not only wired lines but also wireless lines.

[0014] The information terminal 10 is an information processing device such as a smartphone or tablet terminal that wirelessly receives brain waves, which are an example of biosignals measured by an electroencephalograph E in a helmet H worn by a patrolman, determines the patrolman's state of consciousness and emotional state through various analyses of the brain waves, and estimates the possibility of the patrolman committing fraud based on changes in the patrolman's state of consciousness and emotional state. For example, while the patrolman is patrolling, the information terminal 10 wirelessly receives brain waves such as those shown in FIG. 2(a) from the electroencephalograph E, and performs various analyses of the brain waves to determine changes in the patrolman's state of consciousness and the patrolman's emotional state, such as "tension," "anxiety," "relief," or "relaxation."

[0015] Figure 2(a) shows the changes in the emotional state based on brain waves when a patrolman performs appropriate patrol tasks. In this case, when the patrolman enters the guarded space STS, he experiences emotions of "tension" and "anxiety" due to the anticipation of dangers such as an encounter with a thief (intruder) during his upcoming patrol. Then, as the patrolman confirms that there is no danger during his patrol, his "tension" and "anxiety" gradually ease, and he transitions to emotions of "relaxation" and "relief" by the time he leaves the space.

[0016] Figure 2(b) shows the change in the emotional state based on brain waves when a patrolman commits fraudulent acts such as theft of money or valuables while on patrol. In this case, just as in the case of Figure 2(a), when the patrolman enters the guarded space STS, he experiences emotions of "tension" and "anxiety" due to the anticipation of dangers such as encountering suspicious individuals during his upcoming patrol. If the patrolman discovers money or valuables while on patrol, comes up with the idea of ​​stealing them, and commits fraud, his emotions of "tension" and "anxiety" increase due to his awareness that his fraudulent acts will be discovered, and these emotions continue until he leaves the premises.

[0017] The information terminal 10 focuses on the difference in the patrolman's state of consciousness and emotional state when he or she performs appropriate patrol work and when he or she commits fraudulent acts, as described above in Figure 2, and estimates the possibility of fraudulent acts being committed based on the changes in the state of consciousness and emotional state determined from changes in electroencephalograms.

[0018] When the information terminal 10 receives the electroencephalogram from the electroencephalograph E, the reception is not limited to wireless reception, but may be wired reception.

[0019] The management server 20 is a server device that receives detection information from a security terminal ST installed in the guarded space STS via a network N, and transmits security information including the identification information and detection time contained in the detection information to the information terminal 10 via the network N.

[0020] Details of the configuration and operation of the information terminal 10 and management server 20 will be described later. Note that, in Fig. 1, the security system 1 includes the information terminal 10 and the management server 20, but this is not limited to this, and it is also possible to realize the system without including the management server 20 if the information terminal 10 is configured to directly receive detection information from the security terminal ST.

[0021] (Examples of changes in consciousness and emotional states during patrol work) Hereinafter, specific examples of changes in the consciousness state and emotional state of a patrolman during patrol work will be described with reference to FIGS.

[0022] Figure 4 shows an example of changes in consciousness and emotion during patrol work when there is no possibility of fraud. First, with reference to Figure 4, a specific example of changes in the patrol member's consciousness and emotion during patrol work when there is no possibility of fraud will be described.

[0023] In the following description, patrol work is performed along the patrol route indicated by the arrows in the guarded space STS shown in Fig. 3. That is, after entering the building, patrol work proceeds in the following order: the section detected by motion sensor ST-M1, the section detected by motion sensor ST-M2, the section detected by camera ST-C1 (the room where the safe is located), the section detected by motion sensor ST-M3, the section detected by camera ST-C2, the section detected again by motion sensor ST-M2, and the section detected by motion sensor ST-M1, before leaving the building. Here, a section can be considered, in principle, to be a patrol area detected by any sensor, but is not limited to this. It can also include a patrol area not detected by any sensor, such as the section from when camera ST-C2 is no longer detected until motion sensor ST-M2 detects it.

[0024] The patrolman enters the building at 03:49:20. At this time, anticipating the risk of encountering a thief (intruder) during his upcoming patrol, his state of consciousness changes to a "medium" level, and his emotional states of "anxiety," "fear," and "tension" increase to a "medium" level. As the patrol proceeds smoothly and safety within the guarded space STS is confirmed, his emotional states of "anxiety," "fear," and "tension" settle to a "low" level. Later, at 03:56:30, the patrol ends, and once safety within the guarded space STS is confirmed, his state of consciousness returns to normal, resulting in a "medium" level change in his state of consciousness.

[0025] Figure 5 shows an example of changes in consciousness and emotion during patrol work when fraud has been committed and cameras are present. Next, with reference to Figure 5, we will explain a specific example of changes in the patrol member's consciousness and emotion during patrol work when fraud has been committed and cameras are present.

[0026] As in Figure 4, the patrolman enters the building at 03:49:20. Anticipating the possibility of encountering a burglar (intruder) during his upcoming patrol, his state of consciousness changes to a moderate level, and his emotional states of anxiety, fear, and tension increase to a moderate level. Then, assuming that the patrolman decides to steal money and valuables in the area detected by camera ST-C1 (the room where the safe is located) at 03:52:00 and commits the fraudulent act, his state of consciousness changes significantly, and his emotional state of tension also increases to a high level, due to his awareness that his fraudulent act will be discovered. Furthermore, when the patrolman moves to the area detected by another camera, ST-C2, at 03:53:10, the presence of camera ST-C2 causes his state of consciousness to change to a moderate level, and his emotional states of anxiety and tension increase to a high level. After that, the emotional state of the patrolman, which would normally be "anxiety" and "fear" would calm down as he prepares to leave the building, but due to the fear of being exposed for his misconduct, the emotional state continues without calming down.

[0027] Fig. 6 is a diagram showing an example of changes in consciousness and emotion during patrol work when fraud has occurred and there are no cameras. Next, with reference to Fig. 6, a specific example of changes in the consciousness and emotional state of a patrolman during patrol work when fraud has occurred and there are no cameras will be described. Specifically, the description will be given assuming that cameras ST-C1 and ST-C2 are not installed in the guarded space STS shown in Fig. 3.

[0028] As in Figure 4, the patrolman enters the building at 03:49:20. At this time, anticipating the possibility of encountering a burglar (intruder) during his upcoming patrol, his state of consciousness changes to a moderate level, and his emotional states of anxiety, fear, and tension increase to a moderate level. If the patrolman then decides to steal money and valuables in the room where the safe is located (a section without cameras) and commits the fraudulent act, his state of consciousness changes significantly, and his emotional state of tension also increases to a high level, due to his awareness that his fraudulent act will be discovered even if there are no cameras. Furthermore, even if the patrolman moves to another section without cameras, his state of consciousness changes to a moderate level, due to his fear of his fraudulent act being discovered, and his emotional states of anxiety and tension increase to a high level. After that, the patrolman's emotional states, such as "anxiety" and "fear," would normally calm down as he prepared to leave the facility, but due to the fear of being found out for his misconduct, the emotional states continue to fluctuate. Furthermore, his state of consciousness also continues to change without settling down.

[0029] Fig. 7 is a diagram showing an example of changes in consciousness and emotions during patrol work when there is no possibility of fraud and appropriate measures have been taken in response to the cause of the detection information that triggered an alert. Next, with reference to Fig. 7, a specific example of changes in the consciousness and emotional state of a patrol worker during patrol work when there is no possibility of fraud and appropriate measures have been taken in response to the cause of the detection information that triggered an alert will be described. Here, an example will be described in which an alert is issued (detection information is sent to management server 20) due to the detection operation of human presence sensor ST-M2, and the patrol work is carried out with the aim of taking measures in response to the cause of the alert.

[0030] As in Figure 4, the patrolman enters the building at 03:49:20. At this time, anticipating the possibility of encountering a burglar (intruder) during his upcoming patrol, his state of consciousness changes to "moderate," and his emotional states of "anxiety," "fear," and "tension" reach "moderate" levels. Then, in the area detected by the motion sensor ST-M2, the patrolman identifies the cause of the alarm (e.g., a malfunction due to deterioration of the sensor itself or a false detection of a nearby fax machine) and takes appropriate action (e.g., replacing the sensor or adjusting the sensor's sensitivity with a detection test). This brings about a significant change in his state of consciousness, as he feels relieved that he has taken appropriate action. Afterward, the patrolman's emotional states of "anxiety," "fear," and "tension" settle to "low" levels, as he feels relieved that he has identified the cause of the alarm and taken appropriate action. Furthermore, there is no significant change in the state of consciousness, and when the patrol ends at 03:56:30 and the safety of the guarded space STS is confirmed, the state of consciousness returns to normal, resulting in a "moderate" change in the state of consciousness.

[0031] Fig. 8 is a diagram showing an example of changes in consciousness and emotions during patrol work when there is no possibility of fraud and appropriate measures are not taken in response to the cause of the detection information that triggered an alert. Next, with reference to Fig. 8, a specific example of changes in the consciousness and emotional state of a patrol person during patrol work when there is no possibility of fraud and appropriate measures are not taken in response to the cause of the detection information that triggered an alert will be described. Note that the description will be given assuming that an alert is issued (detection information is sent to management server 20) due to the detection operation of human sensor ST-M2, and that the patrol work also serves the purpose of taking measures in response to the cause of the alert.

[0032] As shown in Figure 4, the patrolman enters the building at 03:49:20 and, imagining the danger of encountering a burglar (intruder) during his upcoming patrol, experiences a moderate change in his state of consciousness, resulting in moderate emotional states of anxiety, fear, and tension. Then, in the area detected by the motion sensor ST-M2, the patrolman identifies the cause of the alarm (e.g., malfunction due to deterioration of the sensor itself or a false detection of a nearby fax machine). This identification of the cause of the alarm significantly changes his state of consciousness. Next, the patrolman fails to take appropriate action regarding camera ST-C2 (e.g., identifying the cause of the alarm, checking the camera's shooting range, checking for changes in the shooting environment, etc.). Later, the patrolman reflects on the fact that he did not take appropriate action, resulting in a moderate change in his state of consciousness due to anxiety about the failure to take appropriate action. Furthermore, the emotional state of "anxiety," which would normally calm down as the patient prepares to leave the facility, continues to change to a "moderate" level due to anxiety over the revelation that appropriate measures had not been taken.

[0033] As described above, if a patrol member commits fraudulent acts such as theft of money or valuables during patrol work or if appropriate measures are not taken to address the cause of an alert, the patrol member's state of consciousness and emotional state may change significantly immediately before or during the fraudulent or inappropriate act, or emotional states such as "anxiety" and "tension" may persist while the patrol member's emotional state would normally calm down as the patrol member leaves the building. The security system 1 according to this embodiment detects such changes in the patrol member's state of consciousness and emotional state and estimates the possibility that the patrol member has committed fraud. Note that inappropriate measures taken to address the cause of an alert, as described above, can also be considered an example of fraudulent acts.

[0034] (Hardware configuration of information terminal) 9 is a diagram showing an example of the hardware configuration of an information terminal according to the embodiment, with reference to which the hardware configuration of the information terminal 10 according to the embodiment will be described.

[0035] As shown in FIG. 9, the information terminal 10 includes a CPU (Central Processing Unit) 401, a ROM (Read Only Memory) 402, a RAM (Random Access Memory) 403, an EEPROM (Electrically Erasable Programmable Read Only Memory) 404, an imaging unit 405, an imaging I / F 406, an acceleration / direction sensor 407, and a GPS (Global Positioning System) receiving unit 408.

[0036] The CPU 401 is a computing device that controls the overall operation of the information terminal 10. The ROM 402 is a non-volatile storage device that stores programs used to drive the CPU 401, such as an IPL (Initial Program Loader). The RAM 403 is a volatile storage device that is used as a work area for the CPU 401. The EEPROM 404 is a non-volatile storage device that stores programs and various data.

[0037] The imaging unit 405 is a built-in imaging device that captures an image of a subject using an image sensor such as a CMOS (Complementary Metal Oxide Semiconductor) to obtain image data under the control of the CPU 401. Note that instead of a CMOS image sensor, an image sensor such as a CCD (Charge Coupled Device) may also be used. The imaging I / F 406 is an interface for controlling the driving of the imaging unit 405.

[0038] The acceleration / direction sensor 407 is one of various sensors such as an electronic magnetic compass that detects geomagnetism, a gyrocompass, an acceleration sensor, and the like.

[0039] The GPS receiving unit 408 is a receiving device that receives GPS signals from GPS satellites.

[0040] As shown in FIG. 9, the information terminal 10 further includes a long-distance communication circuit 410, an antenna 410a, a short-distance communication circuit 411, an antenna 411a, a microphone 412, a speaker 413, a sound input / output I / F 414, a display 415, an external device connection I / F 416, a vibrator 417, and a touch panel 418.

[0041] The long-distance communication circuit 410 is a communication circuit that performs wireless communication with other devices via a network N through an antenna 410a in accordance with standards such as Wi-Fi (registered trademark).

[0042] The short-distance communication circuit 411 is a communication circuit that performs short-distance wireless communication with other devices via an antenna 411a in accordance with standards such as NFC (Near Field Communication) or Bluetooth (registered trademark).

[0043] The microphone 412 is a built-in sound collecting device that converts sound into an electrical signal. The speaker 413 is a built-in acoustic device that converts the electrical signal into physical vibrations and outputs sound such as music or voice. The sound input / output I / F 414 is an interface that processes input and output of sound signals between the microphone 412 and the speaker 413 under the control of the CPU 401.

[0044] The display 415 is a display device such as a liquid crystal display or an organic EL (Electro-Luminescence) display that displays an image of a subject, various icons, etc. The external device connection I / F 416 is an interface conforming to standards such as USB (Universal Serial Bus) for connecting various external devices.

[0045] The vibrator 417 is a device that generates physical vibrations under the control of the CPU 401 .

[0046] The touch panel 418 is an input device that allows the user to perform various functions of the information terminal 10 by touching the display 415 .

[0047] The above-mentioned CPU 401, ROM 402, RAM 403, EEPROM 404, imaging I / F 406, acceleration / direction sensor 407, GPS receiver 408, long-distance communication circuit 410, short-distance communication circuit 411, sound input / output I / F 414, display 415, external device connection I / F 416, vibrator 417, and touch panel 418 are connected to each other so as to be able to communicate with each other via bus lines 409 such as an address bus and a data bus.

[0048] The hardware configuration of the information terminal 10 shown in FIG. 9 is an example, and it is not necessary for the information terminal 10 to include all of the components, and other components may also be included.

[0049] (Management server hardware configuration) 10 is a diagram showing an example of the hardware configuration of the management server according to the embodiment, and the hardware configuration of the management server 20 according to the embodiment will be described with reference to FIG.

[0050] As shown in FIG. 10, the management server 20 includes a CPU 501 , a ROM 502 , a RAM 503 , an auxiliary storage device 505 , a network I / F 508 , a display 509 , a keyboard 511 , and a mouse 512 .

[0051] The CPU 501 is a computing device that controls the overall operation of the management server 20. The ROM 502 is a non-volatile storage device that stores programs for the management server 20. The RAM 503 is a volatile storage device that is used as a work area for the CPU 501.

[0052] The auxiliary storage device 505 is a storage device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive) that stores various data, programs, and the like.

[0053] The network I / F 508 is an interface for communicating data with external devices such as the security terminal ST and the information terminal 10 via the network N. The network I / F 508 is, for example, a NIC (Network Interface Card) that is compatible with Ethernet (registered trademark) and capable of communication compliant with TCP / IP or the like.

[0054] The display 509 is a display device configured by a liquid crystal or organic EL display, etc., that displays various information such as a cursor, a menu, a window, characters, or an image.

[0055] The keyboard 511 is an input device for selecting letters, numbers, and various instructions, moving the cursor, etc. The mouse 512 is an input device for selecting and executing various instructions, selecting a processing target, moving the cursor, etc.

[0056] The above-mentioned CPU 501, ROM 502, RAM 503, auxiliary storage device 505, network I / F 508, display 509, keyboard 511, and mouse 512 are communicably connected to one another by a bus line 510 such as an address bus and a data bus.

[0057] The hardware configuration of the management server 20 shown in FIG. 10 is an example, and does not necessarily include all of the components shown in FIG. 5, or may include other components.

[0058] (Configuration and operation of security system functional blocks) FIG. 11 is a diagram showing an example of the configuration of functional blocks of a security system according to an embodiment. FIG. 12 is a diagram showing an example of the distribution of feature amounts based on biological signals. FIG. 13 is a diagram explaining the operation of calculating the amount of change in the level of consciousness in the awareness analysis unit. FIG. 14 is a diagram explaining the distribution of emotional states based on the results of emotional analysis. FIG. 15 is a diagram showing an example of a method of emotion analysis in the emotion analysis unit. FIG. 16 is a diagram showing an example of a method of emotion analysis in the emotion analysis unit. FIG. 17 is a diagram showing an example of operation history information. The configuration and operation of functional blocks of security system 1 according to this embodiment will be described with reference to FIGS. 11 to 17.

[0059] As shown in FIG. 11, the information terminal 10 has a biosignal acquisition unit 101 (first acquisition unit), a feature calculation unit 102 (calculation unit), an awareness analysis unit 103 (an example of an analysis unit), an emotion analysis unit 104 (an example of an analysis unit), an operation history generation unit 105 (generation unit), a security information acquisition unit 106 (second acquisition unit), a fraud estimation unit 107 (estimation unit), a fraud confirmation unit 108 (confirmation unit), a memory unit 109, and an input unit 110.

[0060] The biological signal acquisition unit 101 is a functional unit that acquires biological signals such as brain waves of the patrolman measured by an electroencephalograph E built into a helmet H worn by the patrolman via a short-range communication circuit 411. The biological signal acquisition unit 101 outputs the acquired biological signals such as brain waves to the feature calculation unit 102.

[0061] The feature calculation unit 102 is a functional unit that calculates predetermined feature amounts from biosignals such as electroencephalograms acquired by the biosignal acquisition unit 101. Examples of the predetermined feature amounts include the frequency of electroencephalograms, myoelectric potentials, frequency characteristics obtained by FFT (Fast Fourier Transformation), and frequency components including time information obtained by wavelet transformation. Furthermore, the waveform information of electroencephalograms themselves may be used as the feature amount. The feature calculation unit 102 outputs the calculated feature amounts to the awareness analysis unit 103, the emotion analysis unit 104, and the fraud confirmation unit 108.

[0062] The awareness analysis unit 103 is a functional unit that calculates the degree of awareness of the patrolman by analyzing the feature amounts received from the feature amount calculation unit 102 and calculates the amount of change in the degree of awareness. Here, the degree of awareness is an index value that indicates the state of awareness of the patrolman, such as whether or not he is paying attention to the confirmation work during the patrol, whether or not he is distracted by thinking about something else, or whether or not he is working in a state of awareness accompanied by anxiety or tension.

[0063] For example, when low beta waves (12.75 to 18.5 [Hz]) increase in the frequency components of brain waves as a feature, it is understood that the person is alert and focused on checking while patrolling. When beta waves decrease, it is understood that the person is thinking about something else while patrolling. When alpha waves decrease and irregular beta waves appear, it is understood that the person is in an emotionally unstable state of consciousness. When high beta waves (18 to 29.75 [Hz]) increase, it is understood that the person is in a state of consciousness accompanied by extreme tension or agitation.

[0064] Furthermore, for example, when feature quantity A and feature quantity B are obtained from the feature quantity calculation unit 102, the awareness analysis unit 103 can calculate the change in the awareness state (change in the level of awareness) by plotting them as a two-dimensional distribution as shown in FIG. 12 and determining the change between the average position of the plot group at time t and the average position of the plot group at time t+1. Furthermore, as shown in FIG. 13, the feature quantity calculation unit 102 may determine the level of awareness for each interval and calculate the change in the level of awareness between the intervals. For example, in FIG. 13, the feature quantity calculation unit 102 can calculate the change in the level of awareness (1) between interval (1) and interval (2) from the level of awareness (1) determined in interval (1) and the level of awareness (2) determined in interval (2).

[0065] The awareness analysis unit 103 may also calculate a weighted average of the amount of change in the awareness state by multiplying the amount of change in the awareness state (amount of change in the level of awareness) for each section by the weight for that section, and dividing the accumulated value by the number of sections (number of analyses). In this case, the weight by which the amount of change is multiplied may be increased in sections where, for example, an image sensor or a safe sensor is installed.

[0066] The awareness analysis unit 103 outputs the determined level of awareness or the amount of change in the level of awareness to the operation history generation unit 105. Note that the amount of change in the level of awareness may be calculated by the fraud estimation unit 107, which will be described later. In this case, the awareness analysis unit 103 may output the level of awareness to the operation history generation unit 105. The awareness analysis unit 103 may also output the time when the analysis was performed, together with the determined level of awareness or the amount of change in the level of awareness.

[0067] The emotion analysis unit 104 is a functional unit that calculates arousal and valence as the emotional state by analyzing the feature quantities received from the feature calculation unit 102. Here, arousal indicates a numerical value representing the level of consciousness, and is expressed as a numerical value between, for example, −1 (low consciousness level) and +1 (high consciousness level). Valence indicates a numerical value representing the level of emotion, and is expressed as a numerical value between, for example, −1 (negative emotion) and +1 (positive emotion). As shown in FIG. 14, the emotional state of the traveler can be grasped by plotting the arousal and valence calculated by the emotion analysis unit 104 with arousal on the vertical axis and valence on the horizontal axis. Furthermore, the emotion analysis unit 104 can calculate, for example, the distance between plots at different times in the two-dimensional distribution shown in FIG. 14 as the amount of change in the emotional state.

[0068] Furthermore, as shown in FIG. 15 , the emotion analysis unit 104 can calculate the overall change in the emotional state (here, the change in the emotional level) throughout the patrol work by accumulating the emotional levels calculated for each section. In this case, the emotion analysis unit 104 may calculate the average change in the emotional state by dividing the cumulative value of the emotional levels calculated for each section by the number of sections (number of analyses). For example, if a patrol member performs an appropriate patrol work, as shown in FIG. 15 , the patrol member tends to feel negative emotions in the first half of the patrol work, so the cumulative emotional level is calculated as a negative value. In the second half of the patrol work, the patrol member tends to feel positive emotions as safety is confirmed, so the cumulative emotional level is calculated as a positive value, and the overall change in the emotional level throughout the patrol work is close to zero. On the other hand, if a misconduct is committed during patrol, the patrol member tends to feel negative emotions not only in the first half but also in the second half of the patrol work, so the overall change in the emotional level throughout the patrol work will be a different value from when the patrol work is performed appropriately. Note that, although FIG. 15 shows an example of calculating the overall change in emotional state (here, the change in emotional level) throughout the patrol work, this is not limited to this, and the change in emotional state may also be calculated for each section.

[0069] Furthermore, as shown in FIG. 16 , the emotion analysis unit 104 can calculate the probability of occurrence of each emotion for each fixed interval and calculate the change in the occurrence probability between the fixed intervals as the change in the emotional state. For example, when focusing on the emotion of "anxiety," if a patrol worker performs appropriate patrol work, as shown in FIG. 16 , the probability of occurrence of the emotion of "anxiety" (e.g., 75[%]) is high in the first half of the patrol work, but the probability of occurrence of the emotion of "anxiety" (e.g., 0[%]) is low in the second half of the patrol work, resulting in a large overall change in the probability of occurrence of the emotion of "anxiety" throughout the patrol work (an example of a change in the emotional state). On the other hand, if a misconduct is committed during patrol, the probability of occurrence of the emotion of "anxiety" (e.g., 75[%]) is high in the first half of the patrol work, and the high probability of occurrence of the emotion of "anxiety" continues even in the second half of the patrol work, resulting in a small overall change in the probability of occurrence of the emotion of "anxiety" throughout the patrol work (an example of a change in the emotional state). 16 shows that the fixed interval includes a plurality of intervals, but a single interval may be set as the fixed interval. Also, while FIG. 16 shows an example of calculating the overall change in the probability of occurrence of a specific emotion throughout the patrol work (an example of the change in the emotional state), the present invention is not limited to this, and the change in the probability of occurrence of a specific emotion may be calculated for each interval.

[0070] The emotion analysis unit 104 outputs the determined arousal level and emotionality level, or the amount of change in the emotional state, to the operation history generation unit 105. Note that the amount of change in the emotional state may be calculated by the fraud estimation unit 107, which will be described later. In this case, the emotion analysis unit 104 may output the arousal level and emotionality level to the operation history generation unit 105. The emotion analysis unit 104 may also output the determined arousal level and emotionality level, or the amount of change in the emotional state, as well as the time when the analysis was performed.

[0071] The security information acquisition unit 106 is a functional unit that acquires security information from the management server 20 via the long-distance communication circuit 410. The security information is information including the identification information of the sensor that detected the patrolman, among the various sensors arranged in the security target space STS, and the time of detection, and is information indicating that the patrolman was detected by that sensor. As described above, various sensors such as human sensors, cameras, image sensors, and door sensors are installed in various locations in the security target space STS, and it is possible to detect in which area within the security target space STS a person or the like is present. The security information acquisition unit 106 outputs the acquired security information to the operation history generation unit 105.

[0072] The operation history generating unit 105 is a functional unit that generates operation history information based on the analysis results by the awareness analyzing unit 103 and the emotion analyzing unit 104 and the alarm information acquired by the security information acquiring unit 106 .

[0073] Specifically, the operation history generation unit 105 compares the time at which the electroencephalograms that form the basis of the analysis results received from the consciousness analysis unit 103 and the emotion analysis unit 104 were acquired with the time included in the security information, and generates operation history information by using analysis results and security information that match (or are deemed to match) the time (detection time shown in FIG. 17), the identification information of the sensor included in the security information, and the analysis results (the consciousness analysis results and emotion analysis results shown in FIG. 17). The operation history information shown in FIG. 17 associates, for example, the detection time "03:51:50," the identification information "ST-M2," the consciousness analysis result "Degree of consciousness (2)," and the emotion analysis result "Arousal level (2) / Emotion level (2)." As a result, it can be recognized that the patrolman was detected by the human presence sensor ST-M2 identified by the identification information "ST-M2" at the time "03:51:50," and was patrolling in a consciousness state of degree (2) and in emotional states of arousal (2) and emotionality (2). Note that the example of the operation history information shown in Fig. 17 is an example in which the amount of change in the consciousness state and the amount of change in the emotional state are calculated by the fraud estimation unit 107, and when the amount of change in the consciousness state is calculated by the awareness analysis unit 103 and the amount of change in the emotional state is calculated by the emotion analysis unit 104, the amount of change in the consciousness information may be registered as the awareness analysis result of the operation history information, and the amount of change in the emotional state may be registered as the emotion analysis result.

[0074] Then, the operation history generating unit 105 stores the generated operation history information in the storage unit 109. Note that the operation history generating unit 105 may output the generated operation history information directly to the fraud inferring unit 107.

[0075] The fraud estimation unit 107 is a functional unit that references the behavior history information stored in the storage unit 109 and estimates the possibility that a patrol member patrolling the guarded space STS has committed fraud based on the consciousness analysis results and emotion analysis results of the behavior history information. For example, the fraud estimation unit 107 may estimate the possibility that a patrol member has committed fraud by performing threshold determination on the amount of change in the consciousness state as the consciousness analysis result and the amount of change in the emotional state as the emotion analysis result. In this case, the fraud estimation unit 107 may estimate that a patrol member has committed fraud when the amount of change in the consciousness state exceeds a threshold for the consciousness state and when the amount of change in the emotional state exceeds a threshold for the emotional state. Furthermore, the fraud estimation unit 107 may estimate that a patrol member has committed fraud when at least one of the amount of change in the consciousness state or the amount of change in the emotional state exceeds its respective threshold.

[0076] Note that the fraud inference unit 107 is not limited to referring to the operation history information stored in the storage unit 109, and may directly receive and use the operation history information generated by the operation history generation unit 105. Furthermore, when the level of consciousness is registered as the result of the awareness analysis and the level of arousal and emotionality are registered as the result of the emotion analysis in the operation history information, the fraud inference unit 107 may calculate the amount of change in the state of consciousness between sections from the level of consciousness, calculate the amount of change in the emotional state between sections from the level of arousal and emotionality, and infer the possibility that a fraudulent act has been committed by a patrol member based on these amounts of change.

[0077] The fraud estimation unit 107 outputs the estimation result of the possibility of fraud to the fraud confirmation unit 108.

[0078] Note that the information terminal 10 is not limited to being provided with all of the analysis functions of the above-mentioned awareness analysis unit 103 and emotion analysis unit 104, and may have at least either the awareness analysis unit 103 or the emotion analysis unit 104 to the extent that a certain degree of accuracy in estimating the possibility of fraudulent activity by the fraud estimation unit 107 can be obtained.

[0079] When the fraud estimation unit 107 estimates that there is a high possibility that fraudulent activity has been committed, the fraud confirmation unit 108 is a functional unit that, when asked by the administrator of the monitoring center via telephone, email, chat, or the like, acquires feature amounts based on biological signals such as electroencephalograms of the patrol member from the feature calculation unit 102 and confirms the possibility that fraudulent activity has been committed by the patrol member based on the response of the feature amounts. Specifically, when the fraud estimation unit 107 estimates that there is a high possibility that fraudulent activity has been committed, the fraud confirmation unit 108 notifies the management server 20 of the monitoring center of the estimation result via the long-distance communication circuit 410. Then, upon confirming that the administrator of the monitoring center has received the notification of the estimation result from the management server 20, the administrator calls the information terminal 10 of the patrol member, asks the patrol member a number of questions related to fraudulent activity, and requests answers to the questions. Then, while the visitor is answering the question, the biosignal acquisition unit 101 acquires biosignals such as electroencephalograms (EEGs) measured by an electroencephalograph E worn by the visitor via the short-range communication circuit 411, and the feature calculation unit 102 calculates a feature based on the biosignal. The fraud confirmation unit 108 then confirms (determines) the possibility that the visitor has engaged in fraudulent behavior based on the response of the feature. Note that in this case, the fraud confirmation unit 108 may confirm (determine) the possibility that the visitor has engaged in fraudulent behavior based on the response of the biosignal itself, such as the visitor's EEGs, as the response of the feature. Alternatively, the fraud confirmation unit 108 may confirm (determine) the possibility that the visitor has engaged in fraudulent behavior based on the biosignal itself, rather than the feature.

[0080] Then, the fraud confirmation unit 108 transmits the confirmation result regarding the possibility that fraudulent activity has been committed by the patrol member to the management server 20 via the long-distance communication circuit 410. This enables the administrator of the monitoring center to determine the possibility that fraudulent activity has been committed by the patrol member based on the confirmation result from the fraud confirmation unit 108.

[0081] The storage unit 109 is a functional unit that stores various information such as the operation history information generated by the operation history generation unit 105. Note that the information stored in the storage unit 109 is not limited to the operation history information, and the storage unit 109 may store information generated or acquired by each of the above-mentioned functional units (for example, information on the results of analysis by the awareness analysis unit 103 and the emotion analysis unit 104). The storage unit 109 is realized by the RAM 403 or the EEPROM 404 shown in FIG. 9.

[0082] The input unit 110 is a functional unit that accepts operational input to the information terminal 10. For example, when a patrol person starts or ends a patrol of the security target space STS, the patrol person may input the information to that effect via the input unit 110, and the information may be transmitted to the management server 20 via the long-distance communication circuit 410. The input unit 110 is realized by the touch panel 418 shown in FIG. 9.

[0083] The above-described biosignal acquisition unit 101, feature calculation unit 102, awareness analysis unit 103, emotion analysis unit 104, behavior history generation unit 105, security information acquisition unit 106, fraud estimation unit 107, and fraud confirmation unit 108 are realized by executing a program by CPU 401 shown in Fig. 9. Note that the program may be a native application directly installed on information terminal 10, or may be a web application running on a web browser. Furthermore, at least some of these functional units may be realized by hardware circuits such as a field-programmable gate array (FPGA) or an application-specific integrated circuit (ASIC).

[0084] Furthermore, the functions of each functional unit of the information terminal 10 shown in Fig. 11 are conceptually shown, and the configuration is not limited to this. For example, the multiple functional units illustrated as independent functional units in the information terminal 10 shown in Fig. 11 may be configured as a single functional unit. On the other hand, the function of a single functional unit in the information terminal 10 shown in Fig. 11 may be divided into multiple functional units, and configured as multiple functional units.

[0085] As shown in FIG. 11, the management server 20 includes a detection information acquisition unit 201, a security information transmission unit 202, and a storage unit 203.

[0086] The detection information acquisition unit 201 is a functional unit that, when the presence of a human or the like is detected by various sensors installed in the security space STS, acquires detection information indicating that the presence has been detected from the security terminal ST via the network N and the network I / F 508. The detection information may include, for example, identification information of the security terminal ST that performed the detection. This detection information makes it possible to recognize in real time which area of ​​the security space STS the patrol person is in. The detection information acquisition unit 201 outputs the acquired detection information to the security information transmission unit 202. The detection information acquisition unit 201 may also store the acquired detection information in the storage unit 203.

[0087] Security information transmission unit 202 is a functional unit that generates security information including the identification information of the sensor included in the detection information acquired by detection information acquisition unit 201 and the detection time, and transmits the security information to information terminal 10 via network I / F 508. Note that the detection information may include the time of the moment when security terminal ST made the detection, and security information transmission unit 202 may include the time included in the detection information in the security information. Security information transmission unit 202 may also generate security information using detection information accumulated in memory unit 203.

[0088] The storage unit 203 is a functional unit that stores (accumulates) the detection information acquired by, for example, the detection information acquisition unit 201. The storage unit 203 is realized by the auxiliary storage device 505 shown in FIG.

[0089] The detection information acquisition unit 201 and the security information transmission unit 202 are realized by executing a program by the CPU 501 shown in Fig. 10. Furthermore, at least some of these functional units may be realized by a hardware circuit such as an FPGA or an ASIC.

[0090] Furthermore, the functional units of the management server 20 shown in Fig. 11 are conceptual representations of functions, and are not limited to such a configuration. For example, the multiple functional units illustrated as independent functional units in the management server 20 shown in Fig. 11 may be configured as a single functional unit. On the other hand, the function of a single functional unit in the management server 20 shown in Fig. 11 may be divided into multiple units and configured as multiple functional units.

[0091] (Flow of patrol work by patrol personnel) 18 is a flowchart showing an example of the flow of patrol work when the security system according to the embodiment is used. The flow of patrol work by a patrolman will be described with reference to FIG.

[0092] First, a patrolman wearing a helmet H with an electroencephalograph E built in and carrying an information terminal 10 arrives at a security space STS, which is a site to be patrolled (step S11).

[0093] The patrol person then operates the input unit 110 of the information terminal 10 to input a command to start patrol, which causes the management server 20 to receive a signal via the long-distance communication circuit 410, thereby commencing patrol of the security target space STS (step S12). As part of the patrol work, the patrol person first inspects the perimeter of the security target space STS (step S13). After inspecting the perimeter, the patrol person enters the security target space STS (step S14). By receiving the detection information, the management server 20 can recognize that the patrol person has entered the security target space STS.

[0094] The patrol person then proceeds to inspect the inside of the security space STS (step S15). The security system 1 then analyzes biological signals such as brain waves acquired from the patrol person patrolling the security space STS, and acquires detection information from security terminals ST installed at various locations in the security space STS, while executing a fraud estimation / confirmation process to estimate and confirm the possibility that fraud has been committed by the patrol person in the security space STS (step S16). The flow of the fraud estimation / confirmation process performed by the security system 1 will be described in detail later with reference to FIG. 19.

[0095] When the patrol finishes patrolling the inside of the security target space STS (step S17), the patrol person leaves the security target space STS (step S18). The patrol person also inputs an instruction to end the patrol via the input unit 110 of the information terminal 10, thereby transmitting the instruction to the management server 20 via the long-distance communication circuit 410.

[0096] The patrol work is carried out by the patrol personnel through the flow of steps S11 to S18 described above.

[0097] (Security system fraud estimation and confirmation process flow) Fig. 19 is a flowchart showing an example of the flow of the fraud estimation / confirmation process of the security system according to the embodiment. The flow of the fraud estimation / confirmation process by the security system 1 according to the present embodiment will be described with reference to Fig. 19. The fraud estimation / confirmation process shown in Fig. 19 corresponds to the process of step S16 shown in Fig. 18.

[0098] <Step S161> When a patrol person starts patrolling the guarded space STS, the biosignal acquisition unit 101 of the information terminal 10 starts acquiring biosignals such as brain waves of the patrol person measured by an electroencephalograph E built into a helmet H worn by the patrol person via the short-range communication circuit 411. The timing at which the biosignal acquisition unit 101 starts acquiring biosignals may be, for example, the timing at which the patrol person sends a notice of the start of patrol to the management server 20 and the information terminal 10 receives a response thereto.

[0099] Also, at this timing, the security information acquisition unit 106 of the information terminal 10 begins acquiring security information from the management server 20 via the long-distance communication circuit 410. Specifically, when various sensors installed in the guarded space STS detect a patrol person, they transmit detection information including the identification information of the sensor and the detection time to the management server 20 via the network N. Then, the detection information acquisition unit 201 of the management server 20 acquires detection information indicating the detection from the security terminal ST via the network N and the network I / F 508, and outputs the detection information to the security information transmission unit 202. The security information transmission unit 202 generates security information including the identification information of the sensor included in the detection information acquired by the detection information acquisition unit 201 and the detection time, and transmits the security information to the information terminal 10 via the network I / F 508. This enables the security information acquisition unit 106 of the information terminal 10 to acquire security information. Then, the security information acquisition unit 106 outputs the acquired security information to the operation history generation unit 105.

[0100] Then, the process proceeds to step S162.

[0101] <Step S162> The feature amount calculation unit 102 of the information terminal 10 calculates a predetermined feature amount from the biosignal such as an electroencephalogram acquired by the biosignal acquisition unit 101. Then, the feature amount calculation unit 102 outputs the calculated feature amount to the awareness analysis unit 103, the emotion analysis unit 104, and the fraud confirmation unit 108. Then, the process proceeds to steps S163 and S164. That is, the processes of steps S163 and S164 are executed in parallel.

[0102] <Step S163> The awareness analysis unit 103 of the information terminal 10 determines the level of awareness of the patrolman by analyzing the feature amount received from the feature amount calculation unit 102. Then, the awareness analysis unit 103 outputs the determined level of awareness to the operation history generation unit 105. Note that the awareness analysis unit 103 may also calculate the amount of change in the level of awareness from the determined level of awareness and output it to the operation history generation unit 105. Then, the process proceeds to step S165.

[0103] <Step S164> The emotion analysis unit 104 of the information terminal 10 determines the arousal level and the emotional level as the emotional state by analyzing the feature amounts received from the feature amount calculation unit 102. The emotion analysis unit 104 outputs the determined arousal level and the emotional level to the action history generation unit 105. Note that the emotion analysis unit 104 may also calculate the amount of change in the emotional state from the determined arousal level and the emotional level. Then, the process proceeds to step S165.

[0104] <Step S165> Then, the operation history generating unit 105 of the information terminal 10 generates operation history information based on the analysis results by the awareness analyzing unit 103 and the emotion analyzing unit 104 and the alarm information acquired by the security information acquiring unit 106.

[0105] Specifically, the behavior history generation unit 105 compares the time at which the electroencephalograms that form the basis of the analysis results received from the consciousness analysis unit 103 and the emotion analysis unit 104 were acquired with the time included in the security information, and, using the analysis results and security information that match (or are deemed to match) in time, generates behavior history information by associating the time (detection time), the identification information of the sensor included in the security information, and the analysis results (consciousness analysis results and emotion analysis results).

[0106] Then, the operation history generating unit 105 stores the generated operation history information in the storage unit 109. Then, the process proceeds to step S166.

[0107] <Step S166> The fraud estimation unit 107 of the information terminal 10 refers to the operation history information stored in the storage unit 109 and estimates the possibility that a patrol member patrolling the guarded space STS has committed a fraudulent act based on the awareness analysis results and the emotion analysis results of the operation history information. If the operation history information registers the level of awareness as the awareness analysis result and the arousal level and emotional level as the emotion analysis results, the fraud estimation unit 107 calculates the amount of change in the state of consciousness between sections from the level of awareness, calculates the amount of change in the emotional state between sections from the arousal level and the emotional level, and estimates the possibility that a fraudulent act has been committed by the patrol member based on these amounts of change. The fraud estimation unit 107 outputs the estimation result of the possibility that a fraudulent act has been committed to the fraud confirmation unit 108. Then, the process proceeds to step S167.

[0108] <Step S167> If the fraud estimation unit 107 estimates that there is a high possibility that fraudulent activity has occurred (step S167: Yes), the process proceeds to step S168, and if there is a low possibility that fraudulent activity has occurred (step S167: No), the fraud estimation / confirmation process is terminated.

[0109] <Step S168> When the administrator of the monitoring center requests the patrolman to answer questions related to fraud, the fraud confirmation unit 108 of the information terminal 10 acquires from the feature calculation unit 102 feature amounts based on the patrolman's biological signals, such as brain waves, at the time of answering, and confirms the possibility that the patrolman has committed fraud based on the response of the feature amounts.

[0110] The fraud estimation and confirmation process is carried out by the security system 1 through the flow of steps S161 to S168 described above. More specifically, in the fraud estimation and confirmation process by the security system 1, the processes of steps S161 to S165 are repeatedly carried out from the start to the end of the patrol work of the guarded space STS by the patrol personnel, and at the end of the patrol work, steps S166 to S168 are carried out.

[0111] As described above, in the security system 1 according to this embodiment, the biosignal acquisition unit 101 acquires biosignals such as electroencephalograms from the electroencephalograph E that detects biosignals related to the patrolman, the feature calculation unit 102 calculates predetermined feature amounts from the biosignals acquired by the biosignal acquisition unit 101, and the fraud estimation unit 107 estimates the possibility that a patrolman has committed fraud in the security target space STS that is the target of patrol, based on at least one change in the consciousness state determined by the consciousness analysis unit 103 or the emotional state determined by the emotion analysis unit 104. This makes it possible to estimate the possibility that a fraudulent act has been committed during patrol based on the patrolman's biosignals.

[0112] Furthermore, in the security system 1 according to this embodiment, when the fraud estimation unit 107 estimates that there is a high possibility that a patrolman has committed fraud, the fraud confirmation unit 108 confirms the possibility that the patrolman has committed fraud based on the biosignals acquired by the biosignal acquisition unit 101 during a call between the manager and the patrolman. This makes it possible to more reliably confirm the possibility that the patrolman has committed fraud based on changes in the biosignals acquired during a call between the manager and the patrolman when it is estimated that the patrolman has committed fraud.

[0113] Furthermore, in the security system 1 according to this embodiment, when a patrolman is detected by one or more sensors installed in the security target space STS, the security information acquisition unit 106 acquires security information including at least information about the security terminal ST that detected the patrolman, the operation history generation unit 105 generates operation history information that associates at least the information about the security terminal ST indicated by the security information with the analysis results of at least one of the consciousness analysis unit 103 or the emotion analysis unit 104, and the fraud estimation unit 107 estimates the possibility that the patrolman has committed fraud based on changes in at least one of the consciousness state or the emotional state indicated by the analysis results of the operation history information. This makes it possible to associate the analysis results with each section defined by the detection operation of the security terminal ST and to capture changes in at least one of the consciousness state or the emotional state between sections, thereby improving the accuracy of estimating the possibility that the patrolman has committed fraud.

[0114] (Variation) A security system 1 according to a modification of the above-described embodiment will be described, focusing on the differences from the security system 1 according to the above-described embodiment. In the above-described embodiment, among the functions constituting the fraud estimation / confirmation process, all functions except for the function of receiving detection information from the security terminal ST and transmitting the security information to the information terminal 10 were described as operations realized by the information terminal 10. In this modification, operations realized by a learning model generated by machine learning for the processing by the awareness analysis unit 103, emotion analysis unit 104, action history generation unit 105, and fraud estimation unit 107 will be described. Note that the overall configuration of the security system 1 according to this modification, and the hardware configurations of the information terminal 10 and management server 20 are the same as those described in the above-described embodiment.

[0115] 20 is a diagram showing an example in which some functions of a security system according to a modified example are replaced by processing using a learning model based on machine learning. The configuration and operation of the security system 1 according to this modified example will be described with reference to FIG.

[0116] The management server 20 acquires biosignals such as electroencephalograms from the visitor in advance, creates training data in which the feature values ​​calculated from the biosignals are labeled with the estimated result of the possibility of fraudulent activity, and generates a learning model through machine learning (supervised learning) based on the training data. That is, the learning model is generated using the feature values ​​of the biosignals as explanatory variables and the estimated result of the possibility of fraudulent activity as the objective variable. In this case, well-known algorithms such as SVM (Support Vector Machine), Random Forest, decision tree, neural network, and GBDT (Gradient Boosting Decision Tree) can be used as the machine learning algorithm.

[0117] Although the feature quantities of biosignals such as electroencephalograms are used as explanatory variables, the present invention is not limited to this, and a learning model may be generated using the biosignals themselves as explanatory variables. Furthermore, the learning model is not limited to being generated by the management server 20, and a learning model generated by an external device may be operated on the management server 20.

[0118] The management server 20 receives, via the network N, feature amounts calculated by the feature amount calculation unit 102 from the biosignals acquired by the biosignal acquisition unit 101 in the information terminal 10, inputs the feature amounts as sample data into a learning model, and acquires, as an output, an estimated result of the possibility of fraudulent activity. The management server 20 then transmits the acquired estimated result of the possibility of fraudulent activity to the information terminal 10 via the network N. If the estimated result of the possibility of fraudulent activity indicates a high possibility of fraudulent activity, the fraud confirmation unit 108 of the information terminal 10 acquires, from the feature amount calculation unit 102, feature amounts based on the biosignals, such as electroencephalograms, of the patrolman when answering questions about fraudulent activity requested by the monitoring center administrator, and confirms the possibility of fraudulent activity by the patrolman based on the response of the feature amounts. Other operations are the same as those of the security system 1 according to the above-described embodiment.

[0119] 20, the functions of the awareness analysis unit 103, emotion analysis unit 104, behavior history generation unit 105, and fraud estimation unit 107 of the information terminal 10 according to the above-described embodiment are replaced by a learning model operated on the management server 20. This achieves the same effects as those achieved by the security system 1 according to the above-described embodiment, and also reduces the processing load on the information terminal 10 by replacing some of the functions of the information terminal 10 with the learning model of the management server 20. Furthermore, by improving the accuracy of the learning process for constructing the above-described learning model, the accuracy of the estimation process for the possibility of fraudulent activity can be improved.

[0120] Although the functions of the awareness analysis unit 103, the emotion analysis unit 104, the operation history generation unit 105, and the fraud estimation unit 107 of the information terminal 10 are replaced by the management server 20 using a learning model based on machine learning, this is not limited to this. For example, by replacing only the awareness analysis unit 103 and the emotion analysis unit 104 with the learning model of the management server 20, it is possible to output the level of awareness, arousal level, and emotional level using the learning model, and generate operation history information and estimate the possibility of fraudulent activity on the information terminal 10. Furthermore, this is not limited to using a learning model, and some of the functions of the information terminal 10 may simply be replaced by the management server 20. Furthermore, the functions of the management server 20, i.e., the functions of the detection information acquisition unit 201 and the security information transmission unit 202, may also be replaced by the information terminal 10.

[0121] In the above-described embodiments and modifications, when at least one of the functional units of the information terminal 10 and the management server 20 is realized by executing a program, the program is provided by being pre-installed in a ROM or the like. In the above-described embodiments and modifications, the programs executed by the information terminal 10 and the management server 20 may be provided by being recorded in an installable or executable file format on a computer-readable recording medium such as a CD-ROM (Compact Disc Read Only Memory), a flexible disk (FD), a CD-R (Compact Disc-Recordable), or a DVD (Digital Versatile Disc). In the above-described embodiments and modifications, the programs executed by the information terminal 10 and the management server 20 may be provided by being stored on a computer connected to a network such as the Internet and downloaded via the network. In the above-described embodiments and modifications, the programs executed by the information terminal 10 and the management server 20 may be provided or distributed via a network such as the Internet. Furthermore, in the above-described embodiments and variations, the programs executed by the information terminal 10 and the management server 20 are modularly structured to include at least one of the above-described functional units, and in terms of actual hardware, the CPU 401, CPU 501 reads and executes the programs from the above-described storage devices (e.g., ROM 402, EEPROM 404, auxiliary storage device 505, etc.), thereby loading and generating the above-described functional units onto the main storage device (RAM 403, RAM 503). [Explanation of symbols]

[0122] 1. Security System 10 Information terminals 20 Management Server 101 Biosignal acquisition unit 102 Feature calculation unit 103 Consciousness Analysis Department 104 Emotion Analysis Department 105 Operation history generation unit 106 Security Information Acquisition Department 107 Fraud Estimation Department 108 Fraud confirmation department 109 Storage section 110 Input section 201 Detection information acquisition unit 202 Security Information Transmission Department 203 Storage section 401 CPU 402 ROM 403 RAM 404 EEPROM 405 Imaging unit 406 Imaging I / F 407 Acceleration and orientation sensor 408 GPS receiver 409 Bus Line 410 Telecommunications circuit 410a antenna 411 Near field communication circuit 411a antenna 412 Mike 413 Speaker 414 Sound input / output I / F 415 Display 416 External device connection I / F 417 Vibrator 418 Touch Panel 501 CPU 502 ROM 503 RAM 505 Auxiliary storage 508 Network I / F 509 Display 510 Bus Line 511 keyboard 512 Mouse E. Electroencephalograph H helmet N Network ST Security Terminal STS Guarded Space ST-C1, ST-C2 cameras ST-M1~ST-M3 Human Sensors

Claims

1. a first acquisition unit that acquires a biological signal related to the visitor from a measurement device that detects the biological signal; a calculation unit that calculates a predetermined feature amount from the biological signal acquired by the first acquisition unit; an analysis unit that determines at least one of a consciousness state and an emotional state of the traveler by analyzing the feature amount calculated by the calculation unit; an estimation unit that estimates the possibility that the patrol member has committed a fraudulent act based on the change in at least one of the consciousness state and the emotional state obtained by the analysis unit; A security system equipped with

2. The security system described in claim 1 further comprises a confirmation unit that, when the estimation unit estimates that there is a high possibility that the patrol member has committed fraudulent activity, presents information regarding the fraud to the patrol member and confirms the possibility that the patrol member has committed fraudulent activity based on the biological signal acquired by the first acquisition unit when the information is presented.

3. a second acquisition unit that acquires security information including at least information about the sensor that detected the patrolman when the patrolman is detected by one or more sensors installed in the security target space; a generating unit that generates operation history information that associates at least the information of the sensor indicated by the security information with the analysis result by the analyzing unit; Furthermore, The security system described in claim 1 or 2, wherein the estimation unit estimates the possibility that the patrolman has committed fraudulent acts based on changes in at least one of the consciousness state or the emotional state indicated by the analysis results of the operation history information.

4. The analysis unit an awareness analysis unit for determining the awareness state of the patrolman; an emotion analysis unit that determines the emotional state of the traveler; The security system according to any one of claims 1 to 3, comprising:

5. the awareness analysis unit calculates a level of awareness for each predetermined interval from the feature amount calculated by the calculation unit, and calculates a change in the level of awareness between intervals; The security system according to claim 4 , wherein the estimation unit estimates the possibility that the patrolman has committed a fraudulent act based at least on the amount of change in the level of awareness obtained by the awareness analysis unit.

6. the awareness analysis unit calculates a level of awareness for each predetermined interval from the feature amount calculated by the calculation unit, and calculates a weighted average of the amount of change in the level of awareness for each interval; The security system according to claim 4 , wherein the estimation unit estimates the possibility that the patrolman has committed a fraudulent act based at least on the weighted average calculated by the awareness analysis unit.

7. the emotion analysis unit calculates at least one of an arousal level or an emotional level for each predetermined section as an emotional state from the feature amount calculated by the calculation unit, and determines an amount of change in the emotional state throughout the patrol work of the patrol person; The security system according to claim 4 , wherein the estimation unit estimates the possibility that the patrolman has committed a fraudulent act based at least on the amount of change in the emotional state obtained by the emotion analysis unit.

8. the emotion analysis unit calculates at least one of arousal level or emotional level for each predetermined interval as an emotional state from the feature amount calculated by the calculation unit, and determines an amount of change in the emotional state for each interval; The security system according to claim 4 , wherein the estimation unit estimates the possibility that the patrolman has committed a fraudulent act based at least on the amount of change in the emotional state obtained by the emotion analysis unit.

9. the biological signal is an electroencephalogram of the visitor; The security system according to any one of claims 1 to 8, wherein the measuring device is an electroencephalograph.

10. an acquiring step of acquiring a biological signal related to the visitor from a measuring device that detects the biological signal; a calculation step of calculating a predetermined feature amount from the acquired biological signal; an analyzing step of determining at least one of a consciousness state or an emotional state of the traveler by analyzing the calculated feature amount; an estimation step of estimating a possibility that the visitor has committed a fraudulent act based on the obtained change in at least one of the consciousness state and the emotional state; A security method having the above.

11. On the computer, an acquiring step of acquiring a biological signal related to the visitor from a measuring device that detects the biological signal; a calculation step of calculating a predetermined feature amount from the acquired biological signal; an analyzing step of determining at least one of a consciousness state or an emotional state of the traveler by analyzing the calculated feature amount; an estimation step of estimating a possibility that the visitor has committed a fraudulent act based on the obtained change in at least one of the consciousness state and the emotional state; A program to execute.

Images