JP5759752B2 - Security device and security operation switching method - Google Patents

Description

  The present invention relates to a security device and a security operation switching method for executing a security operation when an abnormality in a monitoring area is detected.

  Conventionally, a security device that performs a security operation when an abnormality in a monitoring area is detected is known. Such a security device has been widely introduced in general households as the interest in security in recent years and the laying cost are reduced.

  As a security device to be installed in ordinary households, a sensor is installed on the site of a building, and when the sensor on the site reacts to an object, it is judged that there is an illegal intruder who attempts to infiltrate the building. There are known devices that perform a security operation that eliminates illegal intruders (such as the operation of an intruder) (hereinafter, such a security operation is referred to as an intruder exclusion operation).

  However, in the security device that performs the intruder exclusion operation when the sensor on the site reacts to the object, the intruder exclusion operation is executed even when the sensor reacts to the housekeeper or the regular visitor, and May cause mental distress. In addition, when the sensor reacts to a small animal other than a human being or an inanimate object, an intruder exclusion operation may be executed, resulting in inconvenience to neighboring residents. For this reason, in this type of security device, when the factor that the sensor is responding to (sensor reaction factor) is an illegal intruder, a mechanism for performing an effective intruder exclusion operation against the illegal intruder is constructed. It is an important issue.

  Regarding the security device that performs the intruder exclusion operation, Patent Document 1 describes the accuracy of presence of an intruder in the monitoring area by “image detection”, “intruder detection”, “intruder detection”, It is described that the determination is made in three stages of “intruder confirmation” and an appropriate intruder exclusion operation is performed according to the determination result.

JP 2006-277739 A

  However, since the technique described in Patent Document 1 is configured to determine the probability that an intruder exists by image recognition processing on image data obtained by imaging a monitoring area, the intrusion is based on a signal from a sensor that detects an object. It cannot be applied to a security device that performs a person exclusion operation.

  The present invention has been made in view of the above, and is capable of accurately estimating a sensor reaction factor and performing an effective guard operation against the estimated sensor reaction factor and a guard operation switching. It aims to provide a method.

In order to solve the above-described problems and achieve the object, the security device according to the present invention is a security device that executes at least one of a plurality of security operations determined in advance when an abnormality is detected in the monitoring area. A plurality of sensors installed in the monitoring region; an acquisition means for acquiring a sensor response signal that takes a binary value indicating whether or not the sensor is reacting from the plurality of sensors ; and time-series data of the sensor response signal. based on, for each of the plurality of objects categories, the confidence factor computing means for object belonging to each object category to compute confidence factor representing the likelihood that a sensor reacting factors before xenon capacitors response signal, the confidence factor And security operation control means for switching the security operation to be executed.

Further, the security operation switching method according to the present invention is a security operation switching method executed in a security device that executes at least one of a plurality of security operations predetermined when an abnormality in the monitoring area is detected, Obtaining a sensor response signal that takes a binary value indicating whether or not the sensor is reacting from a plurality of sensors installed in the monitoring area; and a plurality of object categories based on time-series data of the sensor response signal each, the steps of the object belonging to each object category to compute confidence factor representing the likelihood that a sensor reacting factors before xenon capacitors reaction signal, on the basis of said confidence, the security operations performed And a step of switching.

  According to the present invention, it is possible to accurately estimate a sensor reaction factor and perform an effective guard operation against the estimated sensor reaction factor.

FIG. 1 is a block diagram showing a schematic configuration of the security device. FIG. 2 is a functional block diagram showing a functional configuration inside the control device. FIG. 3 is a main flowchart for explaining the transition of a series of security operations executed in the security device. FIG. 4 is a flowchart showing the flow of processing when the guidance operation is executed. FIG. 5 is a flowchart showing the flow of processing during execution of the warning operation and the physical defense operation. FIG. 6 is a flowchart showing the flow of processing when executing a threatening operation. FIG. 7 is a flowchart showing the flow of processing during execution of an attack operation. FIG. 8 is a model diagram for explaining the outline of the Bayesian network. FIG. 9 is a diagram showing an example of a conditional probability table representing the conditional probability P (B | A) shown in FIG. FIG. 10 is a model diagram for explaining the outline of the dynamic Bayesian network. FIG. 11 is a model diagram showing a basic model in the initial state of the dynamic Bayesian network shown in FIG. FIG. 12 is a model diagram showing a state in which units are added to the basic model shown in FIG. FIG. 13 is a model diagram showing an outline of a judgment model. FIG. 14 is a diagram for explaining how the determination model is expanded by adding units to the basic model. FIG. 15 is a diagram for explaining the position model, and shows a state in which the site of the building that is the monitoring area is divided into a plurality of areas (detection target areas). FIG. 16 is a diagram illustrating a specific example of a conditional probability table expressing the transition of the behavior of an object. FIG. 17 is a diagram illustrating a specific example of a conditional probability table expressing the transition of the position of an object. FIG. 18 is a diagram illustrating a specific example of a conditional probability table expressing the initial position of an object. FIG. 19 is a diagram illustrating a specific example of a conditional probability table expressing the initial behavior of an object. FIG. 20 is a diagram illustrating a sensor model, and is a diagram illustrating a specific example of sensor installation in a residential area to be guarded. FIG. 21 is a diagram illustrating a specific example of a conditional probability table representing a sensor model. FIG. 22 is a diagram illustrating another example of sensor installation in the site of a residence to be guarded. FIG. 23 is a diagram illustrating another example of the conditional probability table representing the sensor model. FIG. 24 is a diagram illustrating a specific example of certainty factor calculation, and is a diagram illustrating an example of a suspicious person's behavior pattern. FIG. 25 is a diagram illustrating a specific example of the certainty factor calculation, and is a diagram illustrating an example of a behavior pattern of a small animal. FIG. 26 is a diagram illustrating a calculation result of the certainty factor in the example of FIG. FIG. 27 is a diagram illustrating a calculation result of the certainty factor in the example of FIG.

  Exemplary embodiments of a security device and a security operation switching method according to the present invention will be explained below in detail with reference to the accompanying drawings.

[Overview]
The security device according to the present embodiment is a security device that executes at least one of a plurality of security operations determined in advance when an abnormality in the monitoring area is detected. Specifically, this security device uses the site of a building used as a residence for ordinary households as a monitoring area, and as a security operation (intruder exclusion operation) to eliminate intruders that have entered the building site , “Guidance operation”, “warning operation”, “intimidation operation” and “attack operation”. The security device includes a sensor reaction signal obtained from a plurality of sensors installed on a site (monitoring area) of a building, and a certainty factor and a certainty factor weighted sum calculated based on time series data of the sensor reaction signal. Based on (details will be described later), it has a function of switching the security operation to be executed among the above security operations.

  Here, the outline | summary of each said guard operation | movement is demonstrated. Each of the security operations described above has two modes, “drive” and “stop”. The “drive” mode is a mode in which a device designated as a device to be used in each security operation is operated, and the “stop” mode is a mode in which the operation of the device is stopped.

  The “guidance operation” is a guard operation that outputs a guidance voice from an audio output device having at least an audible area from the entrance gate of the site to the entrance of the building (hereinafter referred to as an access area). . The guidance voice is a voice message that prompts a householder to release the security or prompts a regular visitor to leave. While the “guidance operation” is being executed, a voice message (guidance voice) is output from the voice output device, for example, “We are currently guarding. Please leave immediately.

  The purpose of the “guidance operation” is to gently notify that the site is guarded. As a result, security can be lifted for householders, retreating for regular visitors, and mental insecurity for illegal intruders. Note that the sound output device used in the “guidance operation” desirably uses a directional speaker whose access area is an audible area. However, a normal speaker having no directivity may be used.

  The “warning operation” is an operation for outputting a warning sound from an audio output device having at least an audible area in the vicinity of an opening of a building (hereinafter referred to as an opening area). The warning voice is a voice message that gives a warning such as being notified to a security company or the like when not leaving. During the execution of the “warning operation”, for example, a voice message (warning voice) such as “Intruding into the security area. If you keep staying, you will be notified to the security company” is output from the voice output device. In the “warning operation”, the limited notification device is operated together with the output of the warning sound from the sound output device. The limited notification device is used to notify the site that an abnormality has occurred on the site, such as a light that is installed so as to affect only humans existing on the site. is there.

  The purpose of the “warning operation” is to give a strong notification that an intruder has entered the guarded site. This suggests that there is a possibility of being reported to a security company, etc., prompting the housekeepers to quickly remove the security, prompting regular visitors to move out quickly, and discovering illegal intruders Can give you a sense of being. Note that the sound output device used in the “warning operation” preferably uses a directional speaker having an opening area as an audible area. However, when a directional speaker is used as an audio output device used in the “guidance operation”, it can be substituted as an audio output device used in the “warning operation”. In this case, when the security operation to be performed is switched from the “guidance operation” to the “warning operation”, the sound output from the directional speaker is switched from the guidance sound to the warning sound.

  The “intimidating action” is an action of outputting intimidating sound from a sound output device having the opening area at least as an audible area. The threatening voice is a voice message that threatens the content such as completion of a report to a security company or the like. During the execution of “intimidation action”, for example, a voice message (intimidation voice) such as “A security company has been notified. A security guard will arrive soon” is output from the voice output device. Further, in the “intimidation operation”, the surrounding notification device is operated together with the output of the intimidation sound from the sound output device. Ambient alarm device is used to notify the inside and outside of the site that an abnormality has occurred on the site, for example, flashing installed to affect not only the site but also the outside of the site. Such as lights and threatening bells.

  There are two purposes for "intimidating action", one is to notify the intruder that a person (security guard, etc.) is coming, and the other is that there is an abnormality in the monitored site. This is to inform the neighboring residents. This makes it possible for illegal intruders to understand that they have been reported to a security company and that they are likely to receive attention from the surroundings. Can be given. Note that the directional speaker used in the “warning operation” can be used as it is as an audio output device used in the “intimidation operation”. In this case, when the security operation to be performed is switched from the “warning operation” to the “intimidation operation”, the sound output from the directional speaker is switched from the warning sound to the intimidation sound. In addition, when a normal speaker with no directivity is used as an audio output device used for “intimidating action”, an intimidating sound equivalent to an intimidating sound or intimidating bell should be output from the speaker at a volume that reaches outside the premises. Thus, it is possible to substitute this speaker as a surrounding notification device.

  "Attack action" is executed in addition to this "threat action" during execution of "threat action", and gives a load to a person existing in an opening area having a high risk of entering the building. This is an action that activates the attack equipment. As the attacking device, a device that restricts the movement of a human existing in the opening area and has an effect of making the person feel uncomfortable. For example, a trapping net, tear gas, flash light, etc. can be considered as examples of attack devices.

  The purpose of the “attack action” is to physically and mentally burden the illegal intruder and make it difficult to continue the intrusion. This makes it difficult for an illegal intruder to stay on the spot and continue the intrusion work, and to abandon the intrusion due to a decrease in the willingness to intrude.

  The security device according to the present embodiment has a “physical defense operation” as a security operation accompanying the “warning operation” in addition to the above-described security operations. The “physical defense operation” is an operation of operating a physical defense device provided in the opening of the building. A physical defense device is a device that physically enhances the blocking state of an opening of a building, such as an electric auxiliary lock or an electric shutter.

  The purpose of the “physical defense operation” is to make intrusion work by illegal intruders difficult by physically strengthening the blocking state of the opening of the building. Thereby, the invasion can be abandoned by extending the staying time outside the building and enhancing the sense of discovery.

[Device configuration]
FIG. 1 is a block diagram illustrating a schematic configuration of a security device according to the present embodiment. As shown in FIG. 1, the security device according to the present embodiment includes a plurality of sensors 1, a sensor interface 2, a control device 3, a device interface 4, and a security operation installed on the site of a building that is a monitoring area. And various devices 5 for executing (intruder exclusion operation). The control device 3 is connected to a monitoring device 6 installed in a security company via a communication network.

  One or more sensors 1 are installed for each detection target area when the site of a building that is a monitoring area is divided into a plurality of detection target areas. That is, in the security device according to the present embodiment, a plurality of sensors 1 are installed on the site of a building that is a monitoring area, and the type of each sensor 1 is a detection target area in which the sensor 1 is installed. Those suitable for object detection are used. For example, sensors that detect the opening and closing of gates at the entrance of a site, passive infrared sensors that use pyroelectric elements, sensors that detect the blocking of infrared beams, and the presence or absence of objects using infrared reflections A sensor or the like that detects the distance can be effectively used as the sensor 1 in the present embodiment.

  The sensor interface 2 is a device for controlling the plurality of sensors 1 and receiving sensor response signals from the sensors 1. The sensor response signal takes two values, “reaction” indicating that the sensor 1 is responding to the object and “no response” indicating that the sensor 1 is not responding to the object. In the security device according to the present embodiment, regardless of whether or not the sensor 1 is reacting to an object, the sensor interface 2 periodically acquires sensor response signals from the plurality of sensors 1 at a predetermined measurement cycle. 3 is supplied.

  The control device 3 comprehensively controls the entire security device according to the present embodiment. In particular, the control device 3 inputs sensor response signals periodically acquired from the plurality of sensors 1 by the sensor interface 2, and uses a time series data of the sensor response signals to calculate a certainty factor and a certainty factor weighted sum described later. calculate. And the control apparatus 3 performs based on the sensor reaction signal periodically acquired from the plurality of sensors 1 by the sensor interface 2 and the certainty factor and the certainty factor weighted sum calculated from the time series data of the sensor reaction signal. The security operation is switched, and a control command for controlling the operation of the various devices 5 is output to the device interface 4.

  In addition, the control device 3 detects an abnormality with respect to the monitoring device 6 installed in the security company when a predetermined condition is satisfied, for example, when the security operation is switched so as to execute the “intimidation operation”. It also has a function of issuing (reporting) a signal that informs the user.

  In accordance with the control command output from the control device 3, the device interface 4 activates the device related to the security operation to be performed among the various devices 5 and stops the other devices.

  Various devices 5 for executing the security operation include the above-described audio output device 5a, limited notification device 5b, surrounding notification device 5c, attack device 5d, and physical defense device 5e. In FIG. 1, the audio output devices used in each of the “guidance operation”, “warning operation”, and “intimidation operation” are collectively referred to as the audio output device 5a. May be used, or some audio output devices may be shared.

  FIG. 2 is a functional block diagram showing a functional configuration inside the control device 3. As shown in FIG. 2, the control device 3 includes a determination model storage unit 31, a certainty factor calculation unit 32, a certainty factor weighted sum calculation unit 33, and a guard operation control unit 34.

  The judgment model storage unit 31 is a database that stores a judgment model that is a model that expresses the causal relationship of the sensor reaction by a probability model using a dynamic Bayesian network.

  The judgment model is a probability model in which for each of a plurality of predetermined object categories that can be a reaction factor of the sensor 1, a conditional probability of behavior transition and a conditional probability of position transition of an object belonging to each object category are determined. A certain behavior model and a sensor model that is a probability model representing the relationship between the position where the object exists and the reactions of the plurality of sensors 1 are included. Moreover, in this embodiment, since the site of the building which is a monitoring area is divided into a plurality of detection target areas, a configuration in which one or more sensors 1 are installed for each detection target area is assumed. A position model representing the positions of a plurality of detection target regions is defined, and this position model is included in the determination model together with the behavior model and the sensor model which are probability models. Details of a specific example of the determination model will be described later.

  The certainty factor calculation unit 32 uses a determination model stored in the determination model storage unit 31 to determine in advance from time-series data of sensor response signals periodically acquired from the plurality of sensors 1 by the sensor interface 2. For each of the plurality of object categories, a certainty factor representing the probability that an object belonging to each object category is a sensor reaction factor is calculated.

  Specifically, every time the sensor response signal is acquired by the sensor interface 2, the certainty factor calculation unit 32 checks the sensor response signal and buffers the sensor response signal in an internal memory or the like. Then, time series data combining the newly obtained sensor response signal and the past sensor response signal is generated as needed, and the time series data of this sensor response signal is stored in the judgment model storage unit 31. Are calculated for each object category. Here, in this embodiment, three types of categories of human, small animal, and inanimate are assumed as the object category. The certainty factor calculation unit 32 temporarily stores the calculation result in an internal memory or the like, and performs recursive calculation using the previous calculation result when calculating the certainty factor in the next processing cycle. The calculation load is reduced. A specific example of the method of calculating the certainty factor for each object category by the certainty factor calculating unit 32 will be described in detail later.

  The certainty factor weighted sum calculating unit 33 is a weighting factor that decreases the certainty factor corresponding to the human category among the certainty factors of each object category calculated in time series by the certainty factor calculating unit 32 as the initial stage of the time series increases. The certainty weighted sum which is a value obtained by weighted addition is calculated.

  As described above, the certainty factor for each object category is calculated by collating the time series data of the sensor response signal with the judgment model. However, since there is little information obtained in the initial stage of the time series, the certainty factor The error is considered to be large. Therefore, in the present embodiment, in order to obtain highly reliable information, the certainty factor weighted sum calculation unit 33 includes, among the certainty factors for each object category calculated by the certainty factor calculation unit 32 in time series, particularly the object category. = For the certainty factor corresponding to a human, the integral of the certainty factor calculated in time series is taken (the sum of the certainty factors is taken every time). At this time, it is considered that the certainty factor in the initial stage of observation is not reliable, and the certainty factor weighted sum is calculated by weighting and adding the initial certainty factor with a low weighting factor. Note that the weighting coefficient is a value that increases as the observation time elapses, and various monotonically increasing functions can be used, but it is considered optimal to use a sigmoid function.

  The security operation control unit 34 includes a sensor reaction signal periodically acquired from the plurality of sensors 1 by the sensor interface 2, a certainty factor for each object category calculated by the certainty factor calculating unit 32, and a certainty factor weighted sum calculating unit. On the basis of the certainty weighted sum calculated by 33, the “drive” mode of each of the above-described guard operations is performed so that the optimum guard operation corresponding to the current situation is executed among the above-described guard operations. A “stop” mode is set, and a control command for controlling the operation of the various devices 5 is output to the device interface 4.

[Specific example of security operation switching]
Here, a specific example of switching of the security operation by the security operation control unit 34 will be described. Note that the switching of the security operation described below is merely an example, and the switching criterion (condition) is not limited to the example described below, and depends on the environment to which the security device is applied. Can be set with various modifications.

  The security device according to the present embodiment has a “security set mode” and a “security release mode”. While the security device is set to the “security set mode”, the plurality of sensors 1 installed in the monitoring area are When any of them reacts to an object, control by the guard operation control unit 34 is started. And according to control by this guard operation control part 34, either of each above-mentioned guard operation is performed. When the security device is set to the “security release mode”, no security operation is performed even if the sensor 1 installed in the monitoring area reacts to an object.

(Guide operation)
The security operation control unit 34, when the sensor that reacts first among the plurality of sensors 1 installed on the site of the building that is the monitoring area is a sensor that detects the opening and closing of the gate, Select “Guiding action”. This is because a regular user assumed as a main target of the “guidance operation” usually enters the site from the gate.

  In addition, the guard operation control unit 34 switches the guard operation according to the following conditions during execution of the “guidance operation”. That is, during the execution of the “guidance operation”, the sensors installed in the access area among the plurality of sensors 1 continue to react, and the certainty weighted sum calculated by the certainty weighted sum calculating unit 33 is When the certainty factor corresponding to the person calculated by the certainty factor calculation unit 32 is higher than the certainty factor corresponding to the object category other than the human being, which is equal to or greater than the threshold Th1 (first threshold value), the guard operation control unit 34 switches the security operation to be performed from “guidance operation” to “warning operation”. This is because the sensor installed in the access area continues to react even though the guidance output is output from the audio output device 5a, and the object staying in the access area is likely to be a human being. This is because there is a high possibility that a trespasser is staying in the access area, and it is considered effective to shift to the “warning action” when the certainty weighted sum exceeds a certain threshold.

  In addition, during execution of the “guidance operation”, a sensor installed outside the access area among the plurality of sensors 1 reacts, and the certainty factor corresponding to the person calculated by the certainty factor calculation unit 32 is human. When it is higher than the certainty factor corresponding to the object category other than, the guard operation control unit 34 switches the guard operation to be performed from “guidance operation” to “warning operation”. This is because, as a result of outputting the guidance voice from the voice output device 5a, the object moves from the access area to another area in the site, and the situation where the moved object is likely to be a human is accessed by an illegal intruder. This is because there is a high possibility of moving to an area other than the area, and it is considered effective to shift to the “warning operation”.

  Further, during the execution of the “guidance operation”, a sensor installed outside the access area among the plurality of sensors 1 reacts, and the certainty factor corresponding to the person calculated by the certainty factor calculation unit 32 is greater than When a state with a higher certainty corresponding to an object category other than human beings continues for a predetermined time or more, the guard action control unit 34 ends all guard actions including “guidance action”. This is because it is considered that it is not necessary to perform a guard operation in a situation where there is a high possibility that an object moved from the access area to another area in the site is a small animal other than a human being or an inanimate object. In addition, although it is set as one of the conditions of the guard operation end that the sensors installed outside the access area are reacting here, even if the sensors installed in the access area are reacting, If the certainty level corresponding to the object category other than the human being is higher than the certainty level corresponding to the person calculated by the certainty level calculation unit 32, the state including all “guidance operations” is continued. The security operation may be terminated.

  In addition, when the state in which none of the plurality of sensors 1 has reacted during the execution of the “guidance operation” continues for a predetermined time or longer, the guard operation control unit 34 includes the “guidance operation”. End all security operations. This is because the situation in which none of the sensors is reacting is likely because the object that was the sensor reaction factor has moved out of the site, and it is considered unnecessary to perform the guard operation.

  In addition, when the security device is switched from the “security set mode” to the “security release mode” during the “guidance operation”, it can be determined that the householder has performed the security release operation. 34 terminates all the guard operations including the “guidance operation”.

  It should be noted that during execution of the “guidance operation”, a sensor in the audible region of the audio output device 5a that is activated in the “guidance operation” (that is, a range in which the effect of the guidance audio from the audio output device 5a reaches) does not react Then, it is useless to operate the audio output device 5a. Therefore, in such a situation, it is desirable that the guard operation control unit 34 interrupts the output of the guidance voice from the voice output device 5a to reduce useless power consumption.

(Warning action)
On the other hand, among the plurality of sensors 1 installed on the site of the building that is the monitoring area, the sensor that reacts first is a sensor other than the sensor that detects opening and closing of the gate, and is calculated by the certainty calculation unit 32. If the certainty level corresponding to the human is higher than the certainty level corresponding to the object category other than the human, the guard action control unit 34 selects “warning action” as the guard action to be executed first. This is because the illegal intruder assumed as the main target of the “warning action” is likely to enter the site from a place other than the gate. In other words, when a sensor other than the one that detects the opening and closing of the gate reacts first and there is a high degree of certainty that it corresponds to a human being, there is a high possibility that an illegal intruder has entered the premises, and the `` warning action '' is immediately performed. This is because it is considered effective.

  When the “warning operation” is selected as the guard operation to be executed, the guard operation control unit 34 causes the “physical defense operation” to be executed along with the “warning operation”. This "physical defense operation" is performed when the security device is switched from "security set mode" to "security release mode" or when the regular unlocking operation of the lock installed at the entrance of the building is confirmed. To finish. The regular unlocking operation of the lock can be confirmed, for example, by providing a sensor that can detect the unlocked state of the lock.

  Further, the guard operation control unit 34 switches the guard operation according to the following conditions during execution of the “warning operation”. That is, during execution of the “warning operation”, the certainty factor corresponding to the human being calculated by the certainty factor calculating unit 32 is higher than the certainty factor corresponding to the object category other than the human being, and among the plurality of sensors 1. The sensor installed in the opening area is reacting, the certainty weighted sum calculated by the certainty factor weighted sum calculating unit 33 is equal to or greater than the threshold Th2 (second threshold value), and the certainty factor calculation is performed. When the certainty factor corresponding to the non-human object category calculated by the unit 32 is equal to or less than a threshold value Th3 (third threshold value) set to a sufficiently small value, the guard operation control unit 34 performs the guard operation to be executed. Is switched from "warning action" to "intimidation action". This is because, despite the fact that a warning sound is output from the sound output device 5a, there is an object that is likely to be a human being in the opening area. This is because there is a high possibility that an attempt to intrude into the site is attempted, and it is considered effective to perform a “deterrence operation” that has a stronger intruder exclusion effect than the “warning operation”.

  In general, as the opening of a building, there is an opening that has a high risk of entering the building (for example, an opening that has a wide opening such as a sweep window and is relatively easy to destroy), and an intrusion. There are low-risk openings (for example, openings such as high windows that are narrow and difficult to break). When the sensor of the opening area where the risk of entering the building is high is reacting and when the sensor of the opening area where the risk of entering the building is low is reacting, The threshold value may be changed. Specifically, the threshold when the sensor of the opening area with a high intrusion risk is reacting is lower than the threshold when the sensor of the opening area with a low intrusion risk is reacting, It is desirable to adopt a mechanism for switching from “warning operation” to “intimidation operation” early.

  Further, during the execution of the “warning operation”, a state in which the certainty factor corresponding to the object category other than the human being is higher than the certainty factor corresponding to the human being calculated by the certainty factor calculating unit 32 continues for a predetermined time or more. In this case, the guard operation control unit 34 terminates all guard operations including the “warning operation”. This is because it is considered that it is not necessary to perform the guard operation in a situation where there is a high possibility that the sensor reaction factor is a small animal other than a human being or an inanimate object.

  In addition, when the state in which none of the plurality of sensors 1 has reacted during the execution of the “warning operation” continues for a predetermined time or more, the guard operation control unit 34 includes the “warning operation”. End all security operations. This is because the situation in which none of the sensors is reacting is likely because the object that was the sensor reaction factor has moved out of the site, and it is considered unnecessary to perform the guard operation.

  In addition, when the security device is switched from the “security set mode” to the “security release mode” during the execution of the “warning operation”, it can be determined that the householder has performed the security release operation. 34 terminates all the guard operations including the “warning operation”.

  During the execution of the “warning operation”, the audio output device 5a operated in the “warning operation” is within the audible range (that is, the range in which the warning sound from the audio output device 5a is effective) or by the limited notification device 5b. In the situation where the sensor in the range where the effect of the abnormality notification reaches is not reacting, it is useless to operate the audio output device 5a and the limited notification device 5b. Therefore, if the sensor in the audible area of the audio output device 5a is not responding, the security operation control unit 34 interrupts the output of the warning sound from the audio output device 5a, and the abnormality notification by the limited notification device 5b. If the sensor in the range where the above effect is exerted does not react, it is desirable to reduce useless power consumption by interrupting the operation of the limited notification device 5b.

(Intimidation action)
The guard operation control unit 34 performs switching from “warning operation” to “intimidation operation” according to the above-described conditions. However, it is desirable that the surrounding notification device 5c used in the “intimidation operation” is operated with a delay from the timing at which the output of the threatening sound from the audio output device 5a is started. That is, as one of the conditions for switching from “warning action” to “intimidation action”, there is a condition that the certainty weighted sum calculated by the certainty degree weighted sum calculating unit 33 is equal to or greater than the threshold value Th2. The two stages of Th2a and Th2b (Th2a <Th2b) are set. When the certainty weighted sum is equal to or greater than the threshold Th2a, the output of the threatening voice from the voice output device 5a is started, and when the certainty weighted sum is equal to or greater than the threshold Th2b, the surroundings are started. The operation of the notification device 5c is started. This is because, considering the influence on the surroundings, it is considered that it is more convenient to operate the devices in order of decreasing influence on the surroundings. In addition, when there are a device that emits light and a device that emits sound as the surrounding notification device 5c, it is desirable to operate in the order of the device that emits light and the device that emits sound in consideration of the magnitude of influence on the surroundings. .

  Further, the guard operation control unit 34 switches the guard operation according to the following conditions during execution of the “intimidation action”. That is, the certainty weighted sum calculated by the certainty weighted sum calculation unit 33 during execution of the “threatening action” is greater than or equal to the threshold Th4 (fourth threshold) greater than the above threshold Th2 (second threshold). In such a case, the “attack action” is executed in addition to the “threat action”. This is because an intimidating sound is output from the sound output device 5a, and an object that is extremely likely to be a trespasser continues to exist in the opening area even though the surrounding notification device 5c is activated. This is because it is considered effective to perform an “attack action” that gives a direct load to illegal intruders.

  Further, when any of the plurality of sensors 1 reacts while the “intimidating action” is executed, but the sensor in the opening area stops responding, the guard action control unit 34 determines the guard action to be executed as “intimidating”. Switch from "Action" to "Warning action". This is because in the situation where the sensor in the opening area stops responding, it can be judged that the risk of illegal intruders entering the building has decreased, and the “warning action”, which has a lower intruder removal effect than the “intimidation action”, can be determined. This is because it is considered effective. If the “attack action” is being executed in addition to the “intimidation action”, the “attack action” is terminated when switching from the “threat action” to the “warning action” is performed.

  Further, during the execution of the “threatening action”, a state in which the certainty factor corresponding to the object category other than the human being is higher than the certainty factor corresponding to the human being calculated by the certainty factor calculating unit 32 continues for a predetermined time or more. In this case, the security operation control unit 34 ends all the security operations including the “intimidation operation”. This is because it is considered that it is not necessary to perform the guard operation in a situation where there is a high possibility that the sensor reaction factor is a small animal other than a human being or an inanimate object.

  In addition, if the state in which none of the plurality of sensors 1 has reacted during the execution of the “intimidation action” continues for a predetermined time or more, the security operation control unit 34 includes the “intimidation action”. End all security operations. This is because the situation in which none of the sensors is reacting is likely because the object that was the sensor reaction factor has moved out of the site, and it is considered unnecessary to perform the guard operation.

  In addition, when the security device is switched from the “security set mode” to the “security release mode” during the execution of the “threat action”, it can be determined that the householder has performed the security release operation. 34 terminates all the security operations including the “intimidation operation”.

  During the execution of the “intimidation operation”, the sound output device 5a operated in the “intimidation operation” is within the audible area (that is, the range where the effect of the intimidation sound from the audio output device 5a reaches), or by the surrounding notification device 5c. It is useless to operate these audio output devices 5a and surrounding notification devices 5c in a situation where the sensors in the range where the effect of the abnormality notification is not reacted. Therefore, if the sensor in the audible area of the audio output device 5a is not responding, the security operation control unit 34 interrupts the output of the threatening sound from the audio output device 5a, and also notifies the abnormality notification by the surrounding notification device 5c. If the sensor in the range where the effect reaches is not reacting, it is desirable to reduce the wasteful power consumption by interrupting the operation of the surrounding notification device 5c. Similarly, during the execution of the “attack operation”, the guard operation control unit 34 is useless by interrupting the operation of the attack device 5d if the sensor within the range that the effect of the attack device 5d reaches does not react. It is desirable to reduce power consumption.

[Specific example of transition of security operation]
Next, a specific example of the transition of the security operation executed by the security device according to the present embodiment will be described with reference to FIGS. FIG. 3 is a main flowchart for explaining a transition of a series of guard operations, FIG. 4 is a flowchart showing a flow of processing when “guidance operation” is executed, and FIG. 5 shows “warning operations” and “physical operations”. FIG. 6 is a flowchart showing the flow of processing when the “defense action” is executed, FIG. 6 is a flowchart showing the flow of processing when the “defense action” is executed, and FIG. 7 shows the flow of processing when the “attack action” is executed. It is a flowchart to show.

(Main flow)
First, with reference to FIG. 3, the outline | summary of operation | movement by the security apparatus concerning this embodiment is demonstrated. When the security device according to the present embodiment is set to the “security set mode”, sensor response signals periodically acquired from the plurality of sensors 1 by the sensor interface 2 are input to the control device 3. The control device 3 monitors the sensor response signal input from the sensor interface 2 and determines whether or not the sensor response signal output from any one of the sensors becomes “reaction”, that is, one of the plurality of sensors 1. It is determined whether or not is reacting (step S1). And the control apparatus 3 starts control by the guard operation control part 34, when any sensor is reacting (step S1: Yes) (step S2).

  When control by the security operation control unit 34 is started, first, it is determined whether or not the sensor determined to be reacting in step S1 is a sensor that detects opening and closing of the gate (step S3). And when the sensor which detects opening and closing of the gate is reacting (step S3: Yes), "guidance | derivation operation | movement" is performed (step S4). On the other hand, when a sensor other than the sensor that detects opening / closing of the gate is reacting (step S3: No), the “warning operation” is executed (step S5).

  During the execution of the “guidance operation” in step S4, the guard operation control unit 34 determines whether or not the condition for shifting to the “warning operation” described above is satisfied (step S6), and shifts to the “warning operation”. When the condition is satisfied (step S6: Yes), the guard operation to be performed is switched from the “guidance operation” to the “warning operation” (step S5). On the other hand, if the condition for ending the guard operation described above is satisfied before the condition for shifting to the “warning operation” is satisfied (step S6: No), the guard operation control unit 34 ends the “guidance operation”. At the same time, its own control is terminated (step S7). At this time, when the condition for terminating the security operation is switching to the “security release mode” (step S8: Yes), a series of operations by the security device are performed along with the termination of the control by the security operation control unit 34. finish. On the other hand, when the condition for terminating the security operation is a condition other than the switching to the “security release mode” (step S8: No), the process returns to step S1 and the monitoring of the sensor reaction signal is continued.

  Further, during the execution of the “warning operation” in step S5, the guard operation control unit 34 determines at any time whether or not the condition for shifting to the “intimidation operation” described above is satisfied (step S9). When the condition for shifting is satisfied (step S9: Yes), the security operation to be performed is switched from the “warning operation” to the “intimidation operation” (step S10). On the other hand, if the above-described condition for terminating the guard operation is established before the condition for shifting to “intimidation action” is established (step S9: No), the guard operation control unit 34 terminates the “warning action”. At the same time, its own control is terminated (step S7). At this time, when the condition for terminating the security operation is switching to the “security release mode” (step S8: Yes), a series of operations by the security device are performed along with the termination of the control by the security operation control unit 34. finish. On the other hand, when the condition for terminating the security operation is a condition other than the switching to the “security release mode” (step S8: No), the process returns to step S1 and the monitoring of the sensor reaction signal is continued.

  Further, during the execution of the “intimidation operation” in step S10, the guard operation control unit 34 determines at any time whether or not the condition for shifting to the “warning operation” described above is satisfied (step S11), and sets the “warning operation”. If the condition for shifting is satisfied (step S11: Yes), the security operation to be executed is switched from “intimidation operation” to “warning operation” (step S5). Further, the security operation control unit 34 determines at any time whether or not the above-described condition for adding the “attack operation” is satisfied during the execution of the “intimidation operation” in step S10 (step S12). If the condition to be added is satisfied (step S12: Yes), “attack action” is executed (step S13). On the other hand, if the condition for ending the above-described guard operation is satisfied before the condition for shifting to the “warning operation” or the condition for adding the “attack operation” is satisfied (step S12: No), the guard operation control is performed. The unit 34 ends the “intimidation operation” and ends its own control (step S7). At this time, when the condition for terminating the security operation is switching to the “security release mode” (step S8: Yes), a series of operations by the security device are performed along with the termination of the control by the security operation control unit 34. finish. On the other hand, when the condition for terminating the security operation is a condition other than the switching to the “security release mode” (step S8: No), the process returns to step S1 and the monitoring of the sensor reaction signal is continued.

  In addition to the “intimidation operation” in step S10, the guard operation control unit 34 also executes the “attack operation” in step S13, whether the condition for shifting to the “warning operation” described above is satisfied. Is determined at any time (step S14), and when the condition for shifting to the “warning operation” is satisfied (step S14: Yes), the security operation to be performed is switched from “intimidation operation” to “warning operation” (step S5). ). On the other hand, if the condition for ending the guard operation described above is satisfied before the condition for shifting to the “warning operation” is satisfied (step S14: No), the guard operation control unit 34 determines that “the threat action” and “ The “attack action” is terminated and the control of itself is terminated (step S7). At this time, when the condition for terminating the security operation is switching to the “security release mode” (step S8: Yes), a series of operations by the security device are performed along with the termination of the control by the security operation control unit 34. finish. On the other hand, when the condition for terminating the security operation is a condition other than the switching to the “security release mode” (step S8: No), the process returns to step S1 and the monitoring of the sensor reaction signal is continued.

(Guide operation)
Next, a specific example of the control by the guard operation control unit 34 when executing the “guidance operation” in step S4 of FIG. 3 will be described with reference to FIG.

  Step S101: When the “guidance operation” is started, the guard operation control unit 34 outputs a control command for operating the audio output device 5a used in the “guidance operation” to the device interface 4, and the audio output device 5a. To output a guidance voice.

  Step S102: The security operation control unit 34 determines whether or not the security device has been switched from the “security set mode” to the “security release mode”. If the security device has not been switched to the “security release mode” (step S102: No), the process proceeds to step S103. On the other hand, when the security device is switched to the “security release mode” (step S102: Yes), the process proceeds to step S7 in FIG. 3 to stop the output of the guidance voice from the audio output device 5a and end the security operation. Let

  Step S103: The guard operation control unit 34 determines whether or not the state in which any of the plurality of sensors 1 is responding is continued. If none of the sensors is reacting (step S103: No), the process proceeds to step S104. On the other hand, when the state in which any one of the plurality of sensors 1 is continuing is continued (step S103: Yes), the process proceeds to step S105.

  Step S104: The security operation control unit 34 determines whether or not a predetermined time has elapsed since none of the sensors has reacted. And if predetermined time has not passed (step S104: No), it will return to step S102 and will repeat subsequent determination. On the other hand, if the predetermined time has elapsed (step S104: Yes), the process proceeds to step S7 in FIG. 3, the output of the guidance voice from the voice output device 5a is stopped, and the security operation is terminated.

  Step S105: The security operation control unit 34 determines whether or not the reacting sensor is a sensor installed in the access area. If the reacting sensor is a sensor installed in the access area (step S105: Yes), the process proceeds to step S106. On the other hand, if the reacting sensor is other than the sensor installed in the access area (step S105: No), the process proceeds to step S108.

  Step S106: The security action control unit 34 determines that the certainty factor weighted sum calculated by the certainty factor weighted sum calculating unit 33 at present (object category = weighted sum of certainty factors corresponding to humans) is a predetermined threshold value Th1 (first step). 1 threshold) or more. If the certainty weighted sum is greater than or equal to the threshold Th1 (step S106: Yes), the process proceeds to step S107. On the other hand, if the certainty weighted sum is less than the threshold Th1 (step S106: No), the process returns to step S102 and the subsequent determination is repeated.

  Step S107: The security operation control unit 34 determines whether or not the object category calculated by the certainty factor calculation unit 32 at present is a certainty factor corresponding to a human being higher than a certainty factor corresponding to another object category. . If the certainty corresponding to the object category = human is higher than the certainty corresponding to another object category (step S107: Yes), the security operation to be executed is switched from the “guidance operation” to the “warning operation”. On the other hand, if the certainty factor corresponding to the object category = human is a value equal to or smaller than the certainty factor corresponding to another object category (step S107: No), the process returns to step S102 and the subsequent determination is repeated.

  Step S108: The security operation control unit 34 determines whether or not the object category calculated by the certainty factor calculation unit 32 at present is a certainty factor corresponding to a person is higher than the certainty factors corresponding to other object categories. . If the certainty corresponding to the object category = human is higher than the certainty corresponding to the other object category (step S108: Yes), the security operation to be performed is switched from the “guidance operation” to the “warning operation”. On the other hand, if the confidence level corresponding to the object category = human is a value equal to or lower than the confidence level corresponding to the other object category (step S108: No), the process proceeds to step S109.

  Step S109: The guard operation control unit 34 determines whether or not a predetermined time has elapsed since the certainty corresponding to the object category = human being equal to or less than the certainty corresponding to another object category. . And if predetermined time has not passed (step S109: No), it will return to step S102 and will repeat subsequent determination. On the other hand, if the predetermined time has elapsed (step S109: Yes), the process proceeds to step S7 in FIG. 3, the output of the guidance voice from the voice output device 5a is stopped, and the security operation is terminated.

(Warning action)
Next, a specific example of control by the guard operation control unit 34 when executing the “warning operation” in step S5 of FIG. 3 will be described with reference to FIG.

  Step S201: When the “warning operation” is started, the guard operation control unit 34 determines whether or not the “warning operation” is started by shifting from the “guidance operation”. If the “warning operation” is not started by the transition from the “guidance operation” (step S201: No), the process proceeds to step S202. On the other hand, if the “warning operation” is started by the transition from the “guidance operation” (step S201: Yes), the process proceeds to step S203. In the case where the “warning operation” is not started by the transition from the “guidance operation”, the opening / closing of the gate among the plurality of sensors 1 is detected after the security device is set to the “security set mode”. This is a case where the security operation by the security device is started from the “warning operation” because other sensors other than the sensor to be reacted first.

  Step S202: The guard operation control unit 34 determines whether or not the object category calculated by the certainty factor calculation unit 32 at present is a certainty factor corresponding to a person is higher than a certainty factor corresponding to another object category. . If the certainty corresponding to the object category = human is higher than the certainty corresponding to the other object category (step S202: Yes), the process proceeds to step S203. On the other hand, if the certainty factor corresponding to the object category = human is a value equal to or smaller than the certainty factor corresponding to the other object category (step S202: No), the process proceeds to step S7 in FIG. 3 and the security operation is terminated.

  Step S203: The guard operation control unit 34 outputs a control command for operating the physical defense device 5e to the device interface 4 to operate the physical defense device 5e.

  Step S204: The guard operation control unit 34 outputs to the device interface 4 a control command for operating the audio output device 5a used in the “warning operation” and a control command for operating the limited notification device 5b. The warning sound is output from the sound output device 5a, and the limited notification device 5b is operated.

  Step S205: The security operation control unit 34 determines whether or not the security device has been switched from the “security set mode” to the “security release mode”. If the security device has not been switched to the “security release mode” (step S205: No), the process proceeds to step S206. On the other hand, when the security device is switched to the “security release mode” (step S205: Yes), the process proceeds to step S7 in FIG. 3 to output a warning voice from the voice output device 5a and the operation of the limited notification device 5b. And the operation | movement of the apparatus 5e for physical defense is stopped, and a guard operation is complete | finished.

  Step S206: The guard operation control unit 34 determines whether or not the state in which any of the plurality of sensors 1 is responding is continued. If no sensor is responding (step S206: No), the process proceeds to step S207. On the other hand, when the state where any one of the plurality of sensors 1 is responding is continued (step S206: Yes), the process proceeds to step S208.

  Step S207: The guard operation control unit 34 determines whether or not a predetermined time has elapsed since any sensor has stopped reacting. And if predetermined time has not passed (step S207: No), it will return to step S205 and will repeat subsequent determination. On the other hand, if the predetermined time has elapsed (step S207: Yes), the process proceeds to step S7 in FIG. 3 to stop the output of the warning sound from the sound output device 5a and the operation of the limited notification device 5b. End.

  Step S208: The guard operation control unit 34 determines whether or not the object category calculated by the certainty factor calculation unit 32 at present is a certainty factor corresponding to a person is higher than the certainty factors corresponding to other object categories. . If the object category = the certainty factor corresponding to the person is equal to or less than the certainty factor corresponding to the other object category (step S208: No), the process proceeds to step S209. On the other hand, if the certainty corresponding to the object category = human is higher than the certainty corresponding to the other object category (step S208: Yes), the process proceeds to step S210.

  Step S209: The guard operation control unit 34 determines whether or not a predetermined time has elapsed after the certainty level corresponding to the object category = human being equal to or less than the certainty level corresponding to another object category. . And if predetermined time has not passed (step S209: No), it returns to step S205 and repeats subsequent determination. On the other hand, if the predetermined time has passed (step S209: Yes), the process proceeds to step S7 in FIG. 3 to stop the output of the warning sound from the sound output device 5a and the operation of the limited notification device 5b, End.

  Step S210: The guard operation control unit 34 determines whether or not the reacting sensor is a sensor installed in the opening area. If the reacting sensor is a sensor installed in the opening area (step S210: Yes), the process proceeds to step S211. On the other hand, if the reacting sensor is other than the sensor installed in the opening area (step S210: No), the process returns to step S205 and the subsequent determination is repeated.

  Step S211: The security operation control unit 34 calculates a certainty factor weighted sum calculated by the certainty factor weighted sum calculating unit 33 at present (object category = weighted sum of certainty factors corresponding to humans) as a predetermined threshold Th2a (first 2) or more. If the certainty weighted sum is greater than or equal to the threshold Th2a (step S211: Yes), the process proceeds to step S212. On the other hand, if the certainty weighted sum is less than the threshold Th2a (step S211: No), the process returns to step S205 and the subsequent determination is repeated.

  Step S212: The security operation control unit 34 determines whether or not the certainty factor corresponding to the object category other than the human being calculated by the certainty factor calculation unit 32 is equal to or less than a predetermined threshold Th3 (third threshold value). Determine whether. If the certainty factor corresponding to the non-human object category is equal to or less than the threshold Th3 (step S212: Yes), the security operation to be performed is switched from the “warning operation” to the “intimidation operation”. On the other hand, if the certainty factor corresponding to the non-human object category exceeds the threshold Th3 (step S212: No), the process returns to step S205 and the subsequent determination is repeated.

(Intimidation action)
Next, a specific example of the control by the guard operation control unit 34 at the time of executing the “intimidation operation” in step S10 of FIG. 3 will be described with reference to FIG.

  Step S301: When “intimidation operation” is started, the security operation control unit 34 outputs a control command for operating the audio output device 5a used in “intimidation operation” to the device interface 4, and the audio output device 5a. To output a threatening voice.

  Step S302: The security operation control unit 34 determines whether or not the security device has been switched from the “security set mode” to the “security release mode”. If the security device is not switched to the “security release mode” (step S302: No), the process proceeds to step S303. On the other hand, when the security device is switched to the “security release mode” (step S302: Yes), the process proceeds to step S7 in FIG. 3 to stop the output of the threatening voice from the voice output device 5a and end the security operation. Let

  Step S303: The guard operation control unit 34 determines whether or not the state in which any of the plurality of sensors 1 is responding is continued. If none of the sensors is reacting (step S303: No), the process proceeds to step S304. On the other hand, when the state in which any one of the plurality of sensors 1 is continuing is continued (step S303: Yes), the process proceeds to step S305.

  Step S304: The security operation control unit 34 determines whether or not a predetermined time has elapsed since any sensor has stopped reacting. And if predetermined time has not passed (step S304: No), it will return to step S302 and will repeat subsequent determination. On the other hand, if the predetermined time has elapsed (step S304: Yes), the process proceeds to step S7 in FIG. 3, and the output of the threatening voice from the voice output device 5a is stopped, and the security operation is terminated.

  Step S305: The security operation control unit 34 determines whether or not the object category calculated by the certainty factor calculating unit 32 at the present time = the certainty factor corresponding to a person is higher than the certainty factors corresponding to other object categories. . If the object category = the certainty factor corresponding to a person is equal to or less than the certainty factor corresponding to another object category (step S305: No), the process proceeds to step S306. On the other hand, if the certainty corresponding to the object category = human is higher than the certainty corresponding to the other object category (step S305: Yes), the process proceeds to step S307.

  Step S306: The guard operation control unit 34 determines whether or not a predetermined time has passed since the certainty level corresponding to the object category = human being equal to or less than the certainty level corresponding to another object category. . And if predetermined time has not passed (step S306: No), it will return to step S302 and will repeat subsequent determination. On the other hand, if the predetermined time has elapsed (step S306: Yes), the process proceeds to step S7 in FIG. 3, the output of the threatening voice from the voice output device 5a is stopped, and the security operation is ended.

  Step S307: The security operation control unit 34 determines whether or not the reacting sensor is a sensor installed in the opening area. If the reacting sensor is a sensor installed in the opening area (step S307: Yes), the process proceeds to step S308. On the other hand, if the reacting sensor is a sensor other than the sensor installed in the opening area (step S307: No), the security operation to be performed is switched from “intimidation operation” to “warning operation”.

  Step S308: The security action control unit 34 sets the certainty factor weighted sum calculated by the certainty factor weighted sum calculating unit 33 (object category = the certainty factor weighted sum corresponding to humans) to a value larger than the threshold Th2a. It is determined whether or not the threshold value Th2b is greater than or equal to the set threshold value Th2b. If the certainty weighted sum is greater than or equal to the threshold Th2b (step S308: Yes), the process proceeds to step S309. On the other hand, if the certainty weighted sum is less than the threshold Th2b (step S308: No), the process returns to step S302 and the subsequent determination is repeated.

  Step S309: The guard operation control unit 34 outputs a control command for operating the surrounding notification device 5c to the device interface 4, and operates the surrounding notification device 5c.

  Step S310: The security operation control unit 34 determines whether or not the security device has been switched from the “security set mode” to the “security release mode”. If the security device has not been switched to the “security release mode” (step S310: No), the process proceeds to step S311. On the other hand, when the security device is switched to the “security release mode” (step S310: Yes), the process proceeds to step S7 in FIG. 3 and the threatening voice output from the voice output device 5a and the operation of the surrounding notification device 5c are performed. Stop and end the security action.

  Step S311: The security operation control unit 34 determines whether or not the state in which any of the plurality of sensors 1 is responding is continued. If none of the sensors is reacting (step S311: No), the process proceeds to step S312. On the other hand, when the state where any one of the plurality of sensors 1 is responding is continued (step S311: Yes), the process proceeds to step S313.

  Step S312: The security operation control unit 34 determines whether or not a predetermined time has elapsed since any sensor has stopped reacting. And if predetermined time has not passed (step S312: No), it will return to step S310 and will repeat subsequent determination. On the other hand, if the predetermined time has elapsed (step S312: Yes), the process proceeds to step S7 in FIG. 3 to stop the output of the threatening sound from the sound output device 5a and the operation of the surrounding notification device 5c, End.

  Step S313: The security operation control unit 34 determines whether or not the object category calculated by the certainty factor calculation unit 32 at the present time = the certainty factor corresponding to a person is higher than the certainty factors corresponding to other object categories. . If the object category = the certainty factor corresponding to the person is equal to or less than the certainty factor corresponding to the other object category (step S313: No), the process proceeds to step S314. On the other hand, if the certainty corresponding to the object category = human is higher than the certainty corresponding to the other object category (step S313: Yes), the process proceeds to step S315.

  Step S314: The guard operation control unit 34 determines whether or not a predetermined time has passed since the certainty level corresponding to the object category = human being equal to or smaller than the certainty level corresponding to another object category. . And if predetermined time has not passed (step S314: No), it will return to step S310 and will repeat subsequent determination. On the other hand, if the predetermined time has passed (step S314: Yes), the process proceeds to step S7 in FIG. 3, and the operation of the threatening voice output from the voice output device 5a and the operation of the surrounding notification device 5c is stopped, and the security operation is performed. End.

  Step S315: The guard operation control unit 34 determines whether or not the reacting sensor is a sensor installed in the opening area. If the reacting sensor is a sensor installed in the opening area (step S315: Yes), the process proceeds to step S316. On the other hand, if the reacting sensor is a sensor other than the sensor installed in the opening area (step S315: No), the security operation to be performed is switched from “intimidation operation” to “warning operation”.

  Step S316: The security operation control unit 34 sets the certainty factor weighted sum calculated by the certainty factor weighted sum calculating unit 33 at present (object category = weighted sum of certainty factors corresponding to humans) to a value larger than the threshold Th2b. It is determined whether or not the threshold value Th4 (fourth threshold value) or more is set. If the certainty weighted sum is greater than or equal to the threshold Th4 (step S316: Yes), the process proceeds to step S317. On the other hand, if the certainty weighted sum is less than the threshold Th4 (step S316: No), the process returns to step S310 and the subsequent determination is repeated.

  Step S317: The guard operation control unit 34 determines whether or not the certainty factor corresponding to the object category other than the human being calculated by the certainty factor calculation unit 32 at present is equal to or less than a predetermined threshold Th3 (third threshold value). Determine whether. If the certainty factor corresponding to the non-human object category is equal to or less than the threshold Th3 (step S317: Yes), “attack action” is added. On the other hand, if the certainty factor corresponding to the non-human object category exceeds the threshold Th3 (step S317: No), the process returns to step S310 and the subsequent determination is repeated.

(Attack action)
Next, a specific example of the control by the guard operation control unit 34 when executing the “attack operation” in step S13 of FIG. 3 will be described with reference to FIG.

  Step S401: When the “attack operation” is started, the guard operation control unit 34 outputs a control command for operating the attack device 5d to the device interface 4 to operate the attack device 5d.

  Step S402: The security operation control unit 34 determines whether or not the security device has been switched from the “security set mode” to the “security release mode”. If the security device is not switched to the “security release mode” (step S402: No), the process proceeds to step S403. On the other hand, when the security device is switched to the “security release mode” (step S402: Yes), the process proceeds to step S7 in FIG. 3 to stop the operation of the threatening voice output from the voice output device 5a and the surrounding notification device 5c. At the same time, the operation of the attacking device 5d is stopped, and the security operation is terminated.

  Step S403: The guard operation control unit 34 determines whether or not the state in which any of the plurality of sensors 1 is responding is continued. If no sensor is responding (step S403: No), the process proceeds to step S404. On the other hand, when the state where any one of the plurality of sensors 1 is reacting is continued (step S403: Yes), the process proceeds to step S405.

  Step S404: The security operation control unit 34 determines whether or not a predetermined time has elapsed since any sensor has stopped reacting. And if predetermined time has not passed (step S404: No), it will return to step S402 and will repeat subsequent determination. On the other hand, if the predetermined time has elapsed (step S404: Yes), the process proceeds to step S7 in FIG. 3, and the attack device 5d is output together with the output of the threatening sound from the sound output device 5a and the operation stop of the surrounding notification device 5c. To stop the security operation.

  Step S405: The guard operation control unit 34 determines whether or not the object category calculated by the certainty factor calculation unit 32 at present is a certainty factor corresponding to a person is higher than the certainty factors corresponding to other object categories. . If the object category = the certainty factor corresponding to the human being is a value equal to or less than the certainty factor corresponding to the other object category (step S405: No), the process proceeds to step S406. On the other hand, if the certainty corresponding to the object category = human is higher than the certainty corresponding to the other object category (step S405: Yes), the process proceeds to step S407.

  Step S406: The guard operation control unit 34 determines whether or not a predetermined time has elapsed after the certainty level corresponding to the object category = human being equal to or less than the certainty level corresponding to another object category. . And if predetermined time has not passed (step S406: No), it will return to step S402 and will repeat subsequent determination. On the other hand, if the predetermined time has elapsed (step S406: Yes), the process proceeds to step S7 in FIG. 3, and the attack device 5d is output together with the output of the threatening sound from the sound output device 5a and the operation stop of the surrounding notification device 5c. To stop the security operation.

  Step S407: The guard operation control unit 34 determines whether or not the reacting sensor is a sensor installed in the opening area. If the reacting sensor is a sensor installed in the opening area (step S407: Yes), the process returns to step S402 and the subsequent determination is repeated. On the other hand, if the reacting sensor is another sensor other than the sensor installed in the opening area (step S407: No), the security operation to be performed is switched to the “warning operation”.

[Calculation of certainty]
As described above, the switching of the guard operation by the guard operation control unit 34 has been described in detail with a specific example. Next, a specific method for calculating the certainty factor used as one of the criteria for determination when switching the guard operation is described. An example will be described in detail.

  In the security device according to the present embodiment, the certainty factor for each object category calculated by the certainty factor calculation unit 32 of the control device 3 is, as described above, the probability that an object belonging to each object category is a sensor reaction factor. Is a value representing Here, in this embodiment, humans, small animals, and inanimate objects are assumed as object categories, and these actions are not uniform. That is, it is not considered that the objects belonging to these object categories take the same action every time, and the moving route, the moving speed, etc. are considered to change every time. Therefore, it is difficult to construct a definitive single model as the behavior model for each object category, and it is necessary to construct a probabilistic behavior model. Therefore, in the security device according to the present embodiment, a probability model using a dynamic Bayesian network is constructed, and the probability that an object belonging to each object category is a sensor reaction factor is determined by probability inference using the dynamic Bayesian network. Calculate the certainty of representing.

  That is, in the security device according to the present embodiment, in order to execute probability inference using a dynamic Bayesian network, action selection criteria (behavior models) for each of various object categories are constructed using probability models, and sensor The relationship between the reaction and the behavior of the object is constructed as a sensor model by a probability model similar to the behavior model. Then, by using a “judgment model” that combines these two probability models and the position model that represents the positions of the plurality of detection target areas described above, the sensor reaction factor is estimated from the sensor reaction signal obtained in time series. I can do it.

(Dynamic Bayesian network)
Below, the outline | summary of the dynamic Bayesian network used for a judgment model is demonstrated first. A dynamic Bayesian network is a time-based expansion of a Bayesian network, which is a probabilistic model for estimating the cause from observed events.

A Bayesian network is a visual representation of a causal relationship between a plurality of elements constituting an event, and the causal relationship is expressed by a conditional probability. In other words, as shown in FIG. 8, the Bayesian network represents each element constituting an event as a node represented by a random variable, connects each node having a causal relationship with a directed side (arrow), and connects the nodes. It is expressed as a model in which the causal relationship is defined by conditional probability. In the example shown in FIG. 8, the causal relationship between the node A and the node B constituting the event is given as the conditional probability P (B | A), and the causal relationship between the node B and the node C is the conditional probability P (C | B). Note that the leading node A has no arrow connected to this node, and the conditional probability cannot be defined, so a prior probability P (A) determined by the random variable of this node itself is given. The model of FIG. 8 expressed by such prior probabilities and conditional probabilities can be expressed by a mathematical formula as shown in the following formula (1).
P (A, B, C) = P (A) P (B | A) P (C | B) (1)

In the Bayesian network as described above, a node that is the starting point of an arrow connected to a certain node is generally called a parent node. Here, when a parent node for a certain node X is represented as Pa (X) and a model having n nodes is represented by a general formula, the following formula (2) is obtained.

  In the Bayesian network, as described above, the causal relationship between each element constituting the event is defined by the conditional probability, but the actual value of each element is generally expressed discretely. In the example shown in FIG. 8, the conditional probability P (B | A) representing the causal relationship between the node A and the node B is expressed by a conditional probability table as shown in FIG. In the conditional probability table of FIG. 9, the probability that node B is “1” when node A is “1” is 0.6, and node B is “0” when node A is “1”. The probability is 0.4, the probability that the node B is “1” when the node A is “0” is 0.1, and the probability that the node B is “0” when the node A is “0” is 0. 9 is shown. In a Bayesian network, a conditional probability table as described above is prepared as a conditional probability between nodes, and a model of the entire network is constructed.

The cause estimation using the Bayesian network is performed by fixing the elements that can be observed and considering the combination of all patterns of unknown elements. For example, in the example illustrated in FIG. 8, when it is observed that the node C is “1”, whether the result that the node C is “1” is due to the node A being “1”. Is estimated under the condition that the node A is “1” and the node B is “1” and the node C is “1”. The sum of the probabilities that node C becomes “1” and node C becomes “1” is calculated. When the above is expressed by a mathematical formula, the following formula (3) is obtained.
P (C = 1 | A = 1) = α (P (A = 1) P (B = 1 | A = 1) P (C = 1 | B = 1) + P (A = 1) P (B = 0 | A = 1) P (C = 1 | B = 0))
_{= Α (Σ B P (A} = 1) P (B | A = 1) P (C = 1 | B)) ··· (3)
Here, α is a normalization constant, and is a constant that performs normalization with the probability of A = 0 when C = 1.

  As described above, in the Bayesian network, the cause probability can be obtained by calculating the sum of the pattern probabilities for all realization values of unknown elements that cannot be observed with fixed observable element values. Estimation is possible.

  The Bayesian network described above is a static model that does not consider the time axis. Therefore, when considering application to a target where a sensor response signal can be obtained from moment to moment like the security device according to the present embodiment, it is necessary to construct a model corresponding to time series. Therefore, a dynamic Bayesian network, which is a model in which the Bayesian network is developed over time, is used.

  As shown in FIG. 10, the dynamic Bayesian network is a model in which one basic model is used as a unit, and units are added every time an observation result is obtained. The estimation method using such a dynamic Bayesian network is basically the same as the estimation method using the Bayesian network described above, and only the number of nodes increases each time an observation result is obtained.

In the example shown in FIG. 10, the model when the first observation result O1 is obtained is as shown in FIG. When the model of FIG. 11 is expressed as the above equation (1), the following equation (4) is obtained.
P (X0, X1, O1) = P (X0) P (X1 | X0) P (O1 | X1) (4)
Here, it is assumed that the state of X1 is estimated. O1 is fixed for the observation result, and the unknown element is X0. When the above equation (3) is used, the estimated value of the state of X1 can be obtained by calculating as the following equation (5).
P (X1 | O1) = _ΣX0P (X0) P (X1 | X0) P (O1 | X1)
= P (O1 | X1) _ΣX0P (X0) P (X1 | X0) (5)

Next, at the time of the second observation, as shown in FIG. 12, the model has X2 and O2 nodes added. When the model shown in FIG. 12 is expressed by mathematical formulas as before, the following formula (6) is obtained.
P (X0, X1, X2, O1, O2) = P (X0) P (X1 | X0) P (O1 | X1) P (X2 | X1) P (O2 | X2) (6)
Here, since the unknown elements are X0 and X1, the estimated value of the state of X2 can be obtained by calculating as in the following equation (7).
P (X2 | O1, O2) = Σ X0 Σ X1 P (X0) P (X1 | X0) P (O1 | X1) P (X2 | X1) P (O2 | X2)
= P (O2 | X2) Σ X1 P (O1 | X1) P (X2 | X1) Σ X0 P (X0) P (X1 | X0) ··· (7)

In a dynamic Bayesian network, each time a new observation result is obtained, a unit is added to the model and the calculation is performed as described above, so that the cause can be estimated from the observation result obtained in time series. However, if units are added from time to time, there is a possibility that the model becomes very large after a certain period of time, and the amount of calculation increases accordingly, which is not practical. Therefore, when causal relation estimation using a dynamic Bayesian network is performed, it is desirable to suppress an increase in calculation amount by performing recursive calculation as shown in the following formula (8).
P (Xt + 1 | O1, O2,..., Ot + 1) = P (Ot + 1 | Xt + 1) _.SIGMA.XtP (Xt + 1 | Xt, O1, O2,..., Ot) P (Xt | O1, O2, ..., Ot)
= P (Ot + 1 | Xt + 1) Σ _XtP (Xt + 1 | Xt) P (Xt | O1, O2,..., Ot) (8)
In this equation (8), the last term is the estimated value at the previous timing. Thus, by performing recursive calculation using the estimated value of the previous timing, it is possible to effectively suppress an increase in the amount of calculation.

(Judgment model)
Next, the determination model used for calculating the certainty factor will be described in more detail. FIG. 13 is a model diagram showing an outline of a judgment model. As shown in FIG. 13, the determination model is a probability model using a dynamic Bayesian network, and is roughly composed of three models: a position model, an action model, and a sensor model. Among these, the behavior model and the sensor model are probability models expressed by a conditional probability table as shown in FIG. In the determination model of FIG. 13, the subscript portion of each node represents the observation time, and “0” represents the initial state. This judgment model is a model (dynamic Bayesian network) that represents time-series changes. As shown in FIG. 14, a part with a common subscript is added as a unit at each observation timing to the basic model in the initial state. It will be done. Hereinafter, the position model, the behavior model, and the sensor model constituting the determination model will be described with specific examples.

(Position model)
The position model is a model that represents the positions of a plurality of detection target regions that are targets of object detection using the sensor 1. In other words, the site structure of the building that serves as a monitoring area is divided into a plurality of abstract areas according to the surrounding situation and the like, and each area is set as a detection target area. The position of each area at this time is expressed as a position model. Specifically, as shown in FIG. 15, for example, the general site structure of a building can be divided into a gate area, a fence area, an access area, a private area, and an opening area. The position of the area is expressed as a position model.

Here, each area of the position model illustrated in FIG. 15 has the following meaning.
Gate area: It is an area where normal access from outside is performed, the gate is normally closed, and opening and closing requires an operation such as pulling a lever.
Fence area: Represents the Fence (障) part.
Access area: This is the area through which a person coming from the outside passes to go to the entrance of the building.
Private area: Represents the space in the back of the garden that is spatially connected to the access area but is off the route from the gate to the entrance.
Opening area: It represents the opening of the building and the space in front of it.

(Behavior model)
The behavior model is a probability model that represents the behavior style of an object belonging to each object category for each of a plurality of predetermined object categories. That is, the behavior model is based on the conditions as shown in FIG. 9 as to what kind of behavior the object belonging to each object category moves within the building site represented by the above position model. It is a model expressed in the form of an attached probability table. In the determination model assumed in the present embodiment, as shown in FIG. 13, the arrows connecting the elements of the behavior model portion are directed toward two types of nodes: object behavior (At) and object position (Xt). . Therefore, the behavior model has two conditional probability tables: a conditional probability table that expresses the transition of the behavior of the objects belonging to each object category, and a conditional probability table that expresses the transition of the position of the object belonging to each object category. Represented by

  Here, before explaining each conditional probability table, an example of specific realization values of the object category, the object action, and the object position which are nodes of the determination model (dynamic Bayesian network) shown in FIG. 13 will be described. .

  The object category is a classification of sensor reaction factors. In this embodiment, there are three categories: human, small animal, and inanimate. Here, inanimate is a category that collectively refers to sensors that may cause sensor misinformation other than small animals, such as flying objects such as sunlight, wind, and garbage bags.

  The object action is a classification of actions that can be taken by the object category. In this embodiment, the action is classified into four actions of disappearance, entry, movement, and stay. Here, “disappearance” means an operation that disappears from outside the site, and refers to a movement that leaves the outside. Moreover, approach means the operation | movement which enters the site inside from the outside. Moreover, movement means the operation | movement which moves within the site. A stay means an operation that hardly moves from the current position in the site.

  As shown in FIG. 15, the object position represents each area in the site expressed in the above position model.

  Next, two conditional probability tables expressing the behavior model will be described. FIG. 16 shows a specific example of a conditional probability table expressing the transition of the behavior of the object, and FIG. 17 shows a specific example of the conditional probability table expressing the transition of the position of the object.

  The conditional probability table that expresses the transition of an object's behavior represents the probability of the next action to be selected under the condition that the object exists at a certain position in the site by the previously selected action. Is. In other words, when the object takes a certain action and is at the current position, it represents the probability of selecting which action to continue. When the conditional probability of such a behavior transition is expressed by an equation, P (action | object, current position, previous behavior) is obtained. By expressing this conditional probability as a conditional probability table as shown in FIG. 9 using the above-mentioned classification of the object behavior and the position model shown in FIG. 15, the behavior of the object as shown in FIG. A conditional probability table representing the transition is obtained.

  A conditional probability table that expresses the transition of the position of an object is a probability that expresses where the current position is, under the condition that the object has taken some action at the position where it existed at the previous timing. It is a thing. In other words, it represents the probability of where the object is after taking an action that is the position where the object was in front. When the conditional probability of such a position transition is expressed by an equation, P (position | object, previous position, previous action) is obtained. By expressing this conditional probability as a conditional probability table as shown in FIG. 9 using the above-mentioned classification of object behavior and the position model shown in FIG. 15, the position of the object as shown in FIG. A conditional probability table representing the transition is obtained.

  By the way, in the determination model of the initial state (basic model shown on the left side of FIG. 14), there is no action or position node at the previous timing, so the conditional probability tables shown in FIGS. 16 and 17 are applied as they are. It is not possible. Therefore, a conditional probability table corresponding to such an initial state, that is, a conditional probability table that is suitable when the sensor 1 reacts first is prepared.

  As for the position in the initial state, where the object that becomes the reaction factor of the sensor 1 first exists in the building site (initial position) can be expressed as a conditional probability P (position | object), which is shown in FIG. 18 is obtained as a conditional probability table of initial position as shown in FIG. In addition, regarding the behavior in the initial state, what kind of behavior the object that is the reaction factor of the sensor 1 takes first (initial behavior) can be expressed as a conditional probability P (action | object, current position), If this is expressed as a conditional probability table as shown in FIG. 9, a conditional probability table of initial behavior as shown in FIG. 19 is obtained.

  In the determination model shown in FIG. 13, since the nodes representing the object category (Ct) are connected by arrows, it is necessary to define a conditional probability for the causal relationship between the nodes representing the object category. However, it is considered that the object does not change while it is moving (for example, a human is not changed to a small animal), so the probability that the previous object and the current object are equal will change by 1.0. The probability should be given as 0.

(Sensor model)
The sensor model is a probabilistic model that represents the relationship between the position where the object exists and the reactions of the plurality of sensors 1. In other words, the sensor model represents how each sensor 1 reacts when an object is present at a certain position. As shown in the sensor model portion of the judgment model shown in FIG. 13, the sensor model is formed by connecting the node of the object position (Xt) and the node of the sensor response signal (Ot) as an observation result with an arrow. Defined by the conditional probability P (sensor | position). The sensor response signal takes a binary value of reaction / no reaction. Hereinafter, the sensor model will be described in more detail with specific examples.

  As shown in FIG. 15, the site of the building that is the monitoring area is divided into five detection target areas: a gate area, an obstacle area, an access area, a private area, and an opening area. Outdoor open / close sensor 1a in the area, two opposed infrared sensors 1b and 1c in the obstacle area, outdoor space sensor 1d in the access area, outdoor space sensor 1e in the private area, and stay detection sensor 1f in the opening area. Each shall be installed. The outdoor open / close sensor 1a is a sensor that detects opening and closing of the gate, and the opposed infrared sensors 1b and 1c are sensors that detect blocking of the infrared beam. The outdoor space sensors 1d and 1e are passive infrared sensors using pyroelectric elements, and the stay detection sensor 1f is a sensor that detects the presence of an object and the distance to the object using infrared reflection. The detection areas of these sensors 1a to 1f are hatched areas in FIG.

  By expressing the conditional probability P (sensor | position) as a conditional probability table as shown in FIG. 9 under the assumption that the sensors 1a to 1f are installed in the site of the house as described above. A conditional probability table of the sensor model as shown in FIG. 21 is obtained. As described above, the sensor response signals obtained from the sensors 1a to 1f are binary values of reaction / no reaction, and the conditional probability table shown in FIG. The conditional probability shown in FIG. 21 expresses which sensor reacts when an object exists at a certain position in the site.

  In the conditional probability table of this sensor model, there is also a probability that an unreacted state in which none of the sensors 1a to 1f reacts even if an object exists in any area (detection target area) in the site. Giving. This is because it is difficult to completely cover each area (detection target area) defined by the position model with the detection areas of the sensors 1a to 1f, and even if an object is present in the detection area, the sensors 1a to 1f. This is because it may not respond by mistake (not reported). Thus, by giving the probability of a non-responsive state to the conditional probability table of the sensor model, it is not necessary to cover the entire area in the site defined by the position model with the detection area of the sensor 1, and the sensor It is possible to cope with the case where 1 does not react by mistake.

  As described above, the determination model used to calculate the certainty factor generates four location probability models by dividing the site of the building area to be monitored into areas, and expresses the behavior model (FIGS. 16 to FIG. 16). 19) and a conditional probability table (FIG. 21) representing the sensor model. The above five conditional probability tables constituting the judgment model are obtained by converting human (intruder), small animal, inanimate behavior patterns and corresponding sensor reaction states into numerical values, which are prepared in advance by the designer. However, by creating these conditional probability tables using empirical know-how related to security, it is possible to construct an optimum judgment model for application to a security device. Hereinafter, an example of a specific creation criterion for the conditional probability table constituting the judgment model suitable for security will be described.

(Human behavior pattern)
First, the following assumptions are made as a basic way of thinking about human behavior patterns.
“Human moves on the ground (does not fly in the sky) and moves from outside the site to the site.
In order to make such an assumption, in the conditional probability table for the initial position shown in FIG. 18, the probability that a person exists in the area inside the site (access area, private area, opening area, etc.) in the initial state is 0. To do.

  Further, since the definition of the object action described above defines that “entry” is to move from outside the site to inside the site, the conditional probability table for the initial action shown in FIG. 19 excludes the gate area and the obstacle area. In this area, the probability that the initial action of a human being is “entry” is set to zero. Similarly, in the conditional probability table of behavior transition shown in FIG. 16, it exists in an area (access area, private area, opening area, etc.) inside the site without depending on the behavior at the previous timing. The probability that a human selects “approach” is zero. Further, in the conditional probability table of position transition shown in FIG. 17, in the position model shown in FIG. 15, it is impossible to exist in front of the entrance immediately after entering (the distance is too far), When the previous action is “entry”, the probability that a person exists in the opening area is set to zero.

  The above is a description of an explicit action pattern regarding the physical behavior of a person based on the definition of the object action and the object position. With respect to the other parts of the conditional probability tables shown in FIGS. 16 to 19, security know-how is applied to human behavior such as intruders. Below, the example is demonstrated easily.

  Intruders tend to dislike being seen from the surroundings. Therefore, there is a tendency to prefer a place / behavior where the prospect is poor, a place / behavior that is not suspicious from the surroundings even if it is seen, and a place / behavior whose purpose is achieved in an extremely short time. Therefore, in the conditional probability table for initial positions shown in FIG. 18, “gate” is set higher than “enclosure” as the probability of human initial position. This is because opening a gate is like a normal visitor and is less likely to be suspicious, and it is considered possible to enter the site in a shorter time than getting over. However, when the gate itself is strong, when the gate is firmly locked, or when the gate is high, the probability of the initial position of the person should be set higher than the gate. May be preferred.

  Similarly, staying around the gate is at risk of being found and suspicious from the surroundings. Therefore, the conditional transition probability table shown in FIG. 16 shows that the gate area does not depend on the behavior at the previous timing. The probability of selecting “move” rather than “stay” is set higher as an action taken by a human being.

  Also, the access area is an area directly connected from the gate, and when the gate is in a lattice shape, the visibility of the access area is improved. Therefore, in the conditional transition probability probability table shown in FIG. 16, the probability that a person will select “stay” is set low in the access area, and the probability that “movement” will be selected is set high. In addition, since the private area is surrounded by obstacles and the risk of being seen from the surroundings is low even when staying, the probability that a person existing in the private area will select “stay” exists in the access area. Set higher than if

  Humans go inside the building and go through the opening to enter the building. Therefore, the human heads for the opening area. Therefore, the probability that a person will select “move” is generally high in areas other than the opening area, and conversely, the probability that a person will select “stay” is high in the opening area. The conditional probability table for position transition shown in FIG. 17 represents the directionality of the movement of the object. While the intruder moves in an area where the line of sight is poor, the position changes so as to go to the area of the opening where the line of sight is poor and easy to enter. For this reason, in the conditional probability table for position transition shown in FIG. 17, the transition probability of the human position is given so as to proceed to the area preferred by the intruder.

  Conditional probabilities relating to human positions and actions are determined according to the above criteria, but these are adjusted depending on the physical structure of the house to be guarded and the surrounding environment. For example, if the fence is about 1 m, it may take a short time to get over. In such a building, the initial position of the human in the condition probability table for the initial position shown in FIG. To increase the probability of

(Small animal behavior patterns)
As for humans, as in humans, it is assumed that “move the ground (do not fly in the sky) and move from outside the site to the site. The ground is also moved within the site”. In order to make such an assumption, in the conditional probability table of the initial position shown in FIG. 18, the probability that the initial position of the small animal is “enclosure” is set high. In addition, since it is thought that a small animal does not open and close a gate, the probability that it exists in a gate area is set to 0.

  In the other conditional probability tables, items regarding small animals in each conditional probability table are filled in on the assumption that “small animals randomly select an action and move without a direction”.

(Inanimate pattern)
Inanimate objects include natural factors such as wind and sunlight, as well as flying objects such as garbage bags. Therefore, any position in the site can be an inanimate initial position, but the probability depends on the size of each area. For this reason, in the conditional probability table of the initial position shown in FIG. 18, the magnitude of the probability of the inanimate initial position is set according to the size of each area.

  Further, since it is considered that an inanimate object rarely moves, the probability of selecting “move” gives a very small probability in any state (position, behavior at the previous timing). In addition, since inanimate factors may disappear (sunlight is covered with clouds or the sun is tilted), the probability of selecting “disappear” as an action is set high. In accordance with the above concept, the inanimate items in the conditional probability tables shown in FIGS.

(Sensor response signal pattern)
The sensor response signal does not change depending on the behavior pattern of the object, but the probability of reaction changes depending on the type of sensor 1, the size of the detection area, and the installation position of sensor 1. The conditional probability table of the sensor model shown in FIG. 21 shows which sensor reacts when the object exists at a certain position X. In the example shown in FIG. The probability that a sensor other than the sensor installed in an area will react is set to zero.

  In the conditional probability table of the sensor model shown in FIG. 21, as described above, the probability that an object is present but does not react erroneously as the probability of “no reaction” depends on the type of sensor 1. Giving. In the example shown in FIG. 20, the outdoor open / close sensor 1a installed in the gate area, the opposed infrared sensors 1b and 1c installed in the enclosed area, and the stay detection sensor 1f installed in the opening area may not react by mistake. Since it is assumed that there is very little, the probability of no reaction is set to a very small value. On the contrary, the outdoor space sensors 1d and 1e installed in the access area and the private area assume that the probability of not reacting erroneously is high to some extent, so the probability of no reaction is set higher than that of other sensors.

  By the way, in the example shown in FIG. 20, it is assumed that each area (detection target area) defined by the position model is almost covered by the detection areas of the sensors 1a to 1f. It is often difficult to completely cover all areas (detection target areas) with one detection area. Therefore, in such a case, the conditional probability table of the sensor model is also considered in consideration of the possibility that an object exists in the area defined by the position model but outside the detection area of the sensor 1. Determine the probability of “no reaction”.

  For example, as shown in FIG. 22, an outdoor space sensor 1g having a part of the access area as a detection area, a part of the access area, and a private area as shown in FIG. It is assumed that an outdoor space sensor 1h having a part of the detection area as a detection area and an outdoor space sensor 1i having a part of the private area as a detection area are installed (the hatched areas in the figure are each sensor 1g). -1i detection areas are shown). In the example shown in FIG. 22, the area deviating from the sensor detection area is wide particularly in the private area, and therefore there is a high probability that the object will not react even if an object exists in the private area. Therefore, in the conditional probability table of the sensor model, when the position is a private area, the probability is set low according to the area missing from the sensor detection area. FIG. 23 shows an example of a conditional probability table of the sensor model in such a case. For simplicity, only the access area and private area are described here.

  In the example shown in FIG. 22, the outdoor space sensor 1h is installed across the access area and the private area. In this case, the outdoor space sensor 1h may react even if an object exists in either the access area or the private area. Therefore, in the conditional probability table of the sensor model shown in FIG. 23, the probability of the outdoor space sensor 1h is not set to 0 regardless of whether the position of the object is an access area or a private area, and how many sensor detection areas are in each area. Gives a probability depending on whether or not

(Calculation of confidence)
Next, in the certainty factor calculation unit 32 of the control device 3, a method of calculating the certainty factor that is a sensor reaction factor for each object category using the time series data of the reaction signal of the sensor 1 and the determination model described above. Will be described.

The calculation formula for cause estimation using the Bayesian network is composed of a product and a sum of conditional probabilities as shown in the above formula (3). When the above equation (3) is applied to the determination model of FIG. 13 in order to calculate confidence using the determination model as shown in FIG. 13 in the sensor reaction factor estimation algorithm, the following equation (9 )become that way.
P (C0 = c | O0 = o) = α (P (C0 = c) Σ X Σ A P (A0 | C0 = c, X0) P (X0 | C0 = c) P (O0 = o | X0)) ... (9)
Moreover, the calculation formula of the certainty factor when the observed value after the second time is obtained is expressed as a calculation formula for performing recursive calculation as in the following formula (10).
P (C = c | Ot = ot, Ot-1 = ot-1, Ot-2 = ot-2,...)
_{= Α ((Σ Xt Σ At} Σ Xt-1 Σ At-1 P (At | C = c, Xt, At-1) P (Xt | C = c, At-1, Xt-1) P (Ot = ot | Xt)) P (C = c | Ot-1 = ot-1, Ot-2 = ot-2,...)) (10)
In the above formulas (9) and (10), C represents an object category, and O represents a sensor response signal (reacted sensor). A represents the action of the object, and X represents the position where the object exists. The subscript of each variable represents the observation timing, and the subscript “0” represents the initial state. Further, Σ _A (...) Means summing with respect to the action A, and Σ _X (...) Means summing with respect to the position X.

  In the above equation (9), the conditional probability at the initial position is referred to by referring to the value from the conditional probability table as shown in FIG. 19 for P (A0 | C0 = c, X0) representing the conditional probability of the initial action. With respect to P (X0 | C0 = c) representing the value, a value is referred to from the conditional probability table as shown in FIG. 18, and P (O0 = o | X0) representing the conditional probability of the sensor model is shown in FIG. By referring to the value from the conditional probability table, the certainty P (C0 = c | O0 = o) that the object category of the reaction factor is c when o is observed as the sensor response signal in the initial state. Can be calculated.

  Further, in the above equation (10), a value is referred to from the conditional probability table as shown in FIG. 16 with respect to P (At | C = c, Xt, At-1) representing the conditional probability of action transition, With respect to P (Xt | C = c, At-1, Xt-1) representing the conditional probability of the position transition, a value is referred to from the conditional probability table as shown in FIG. 17 to represent the conditional probability of the sensor model. With respect to P (Ot = ot | Xt), a value is referred from the conditional probability table as shown in FIG. 21, and the previous calculation result P (C = c | Ot-1 = ot-1, Ot-2 = ot- 2)), when o is observed as the second and subsequent sensor reaction signals, the certainty factor P (C = c) that the object category of the reaction factor is c | Ot = ot, Ot-1 = ot-1, Ot-2 = ot-2,.

上記の式（１１）から分かるように、知りたい物体カテゴリと観測で得たセンサ反応の部分は固定し、行動Ａと位置Ｘの組み合わせを変えた全てのパターンの確率の総和をとっている。それぞれの条件付確率は、図１６〜図１９、図２１に示したような条件付確率表に照らし合わせることで即座に具体的な数値が割り当てられ、それらの数値を使って確信度を算出する。 Here, in the example shown in FIG. 20, when the response of the outdoor open / close sensor 1a installed in the gate area is observed as the sensor response signal in the initial state, the certainty factor that the sensor response factor is human is calculated. If the above equation (9) is embodied with the assumption of the above, the following equation (11) is obtained.

As can be seen from the above equation (11), the object category desired to be known and the sensor response portion obtained by observation are fixed, and the sum of the probabilities of all patterns with different combinations of action A and position X is taken. Each conditional probability is immediately assigned a specific numerical value by comparing with the conditional probability table as shown in FIGS. 16 to 19 and FIG. 21, and the certainty is calculated using those numerical values. .

  In the above equations (9) to (11), α at the beginning of the right side is a normalization constant, and is calculated when c is changed in three object categories: human, small animal, inanimate. The coefficient is such that the sum of two values is 1.0. In actual calculation, the certainty factor is calculated for each of the three object categories using the above formula without considering α, and after the three values are obtained, each value is divided by the sum of the three values. To gain confidence.

(Specific example of confidence calculation)
Next, a specific example of calculating the certainty factor for each object category assuming the behavior patterns of the objects shown in FIGS. 24 and 25 will be described. FIG. 24 shows an example of a case where a suspicious person enters the access area from the obstacle area and goes directly to the opening area (entrance). It is assumed that the sensor has reacted three times with {enclosure area, access area, opening area}. On the other hand, FIG. 25 shows an example in which a small animal gets over the obstacle and enters the access area as in FIG. The sensor response in this case is {enclosure area, access area, enclosure area}.

  The certainty factor for each object category is calculated by using the above equation (9) and the conditional probability tables as shown in FIGS. 18, 19, and 21 for the first sensor response. From the second sensor response, calculation is recursively using the above equation (10) and the conditional probability tables as shown in FIGS. 16, 17, and 21 so that the calculation amount does not increase.

  FIG. 26 shows the result when the certainty factor is calculated for each object category in the example of FIG. FIG. 27 shows the result when the certainty factor is calculated for each object category in the example of FIG. In these FIG. 26 and FIG. 27, the real part represents the estimated value (confidence) of each sensor reaction factor. As shown in FIG. 26, in the case of the example of FIG. 24, it starts from the reaction of the sensor in the surrounding area facing the road, and is initially indistinguishable from the animal. However, the fact that the directionality goes straight to the opening area (entrance) has been found (that is, the sensor response O2 has been obtained), and the confidence factor that the sensor reaction factor is human is high. It becomes. On the other hand, as shown in FIG. 27, in the case of the example of FIG. 25, the reaction starts from the sensor in the surrounding area as in the example of FIG. . By obtaining a sensor response sequence that does not have a specific purpose, a certainty factor that a small animal is a sensor reaction factor as a moving object without a purpose has a high value.

[Effect of the embodiment]
As described above in detail with reference to specific examples, in the security device according to the present embodiment, the certainty factor calculation unit 32 of the control device 3 periodically acquires sensor response signals from the plurality of sensors 1. The time series data is compared with the judgment model stored in the judgment model storage unit 31, and for each of the plurality of object categories, a certainty factor representing the probability that an object belonging to each object category is a sensor reaction factor is calculated. To do. Further, the certainty factor weighted sum calculating unit 33 of the control device 3 weights the certainty factor corresponding to the object category = human being calculated in time series by the certainty factor calculating unit 32 with a lower weighting factor in the initial stage of the time series. The added confidence weighted sum is calculated. Then, the guard operation control unit 34 of the control device 3 uses the sensor reaction signals obtained from the plurality of sensors 1, the certainty factor for each object category calculated by the certainty factor calculating unit 32, and the certainty factor weighted sum calculating unit 33. Based on the calculated certainty weighted sum, the guard operation is switched so that the optimum guard operation corresponding to the current situation is executed. Therefore, according to the security device according to the present embodiment, the sensor reaction factor can be accurately estimated, and an effective security operation can be performed on the estimated sensor response factor.

  As described above, the security device and the security operation switching method according to the present invention are useful as a technique for performing an effective intruder exclusion operation for illegal intruders.

DESCRIPTION OF SYMBOLS 1 Sensor 3 Control apparatus 5a Audio | voice output apparatus 5b Limited notification apparatus 5c Ambient notification apparatus 5d Attack apparatus 5e Physical defense apparatus 31 Judgment model memory | storage part 32 Certainty factor calculation part 33 Certainty factor weighted sum calculation part 34 Security operation control Part

Claims (12)

In a security device that executes at least one of a plurality of predetermined security operations when an abnormality in the monitoring area is detected,
A plurality of sensors installed in the monitoring area;
An acquisition means for acquiring a sensor response signal that takes a binary value indicating whether the sensor is responding from the plurality of sensors ;
Based on the time-series data of the sensor response signal, and a plurality of each object category, confidence factor computing means for object belonging to each object category to compute confidence factor representing the likelihood that a sensor reaction factor,
Before the xenon capacitors reaction signal, on the basis of said confidence, security apparatus characterized by comprising a security operation control means for switching the security operation to be executed. For each of the plurality of object categories, an action model that is a probability model that defines a conditional probability of action transition of an object belonging to each object category and a conditional probability of position transition; a position where the object exists; A judgment model storage means for storing a judgment model including a sensor model that is a probability model representing a relationship with a sensor response of
The certainty factor calculating means obtains, for each of the plurality of object categories, a probability of taking an action pattern that matches the time series data of the sensor response signal using the action model, and represents the probability with the sensor model. The security device according to claim 1, wherein the certainty factor for each of the plurality of object categories is calculated by multiplying a probability of a sensor reaction with respect to a position of the object to be detected. The object category includes a category indicating a person,
A certainty factor weighted sum calculating unit that calculates a certainty factor weighted sum obtained by weighting and adding the certainty factor corresponding to a human being calculated in a time series by the certainty factor calculating unit using a weighting factor that is lower in the initial stage of the time series. Prepared,
The security operation control means switches a security operation to be executed based on a sensor response signal obtained from the plurality of sensors, the certainty factor, and the certainty factor weighted sum. Security device according to. The site of the building is the monitoring area,
The plurality of security operations are:
A guidance operation for outputting a guidance voice from a voice output device having at least an audible area as an access area that is a route from the entrance of the site to the entrance of the building;
A warning operation for outputting a warning sound from a sound output device having at least an audible area near the opening of the building;
A threatening action of outputting a threatening sound from a sound output device having at least an audible area near the opening of the building;
The security device according to claim 3, further comprising: an attack operation that activates an attack device that applies a load to a person existing near the opening of the building.   The security operation control means responds to a human being when a sensor installed in the access area is reacting during execution of the guidance operation and the certainty weighted sum is equal to or greater than a first threshold value. 5. The security device according to claim 4, wherein when the certainty factor to be performed is higher than the certainty factor corresponding to an object category other than a human being, a security operation to be performed is switched from the guidance operation to the warning operation. .   The security operation control means is installed in the vicinity of the opening of the building, and the certainty level corresponding to a person is higher than the certainty level corresponding to an object category other than a human being during execution of the warning action. This is executed when the sensor is responding, the certainty weighted sum is equal to or greater than the second threshold, and the certainty corresponding to the object category other than human is equal to or smaller than the third threshold. The security device according to claim 4, wherein the security operation is switched from the warning operation to the threatening operation.   The security operation control means performs the attack operation in addition to the threatening operation when the certainty weighted sum is equal to or larger than a fourth threshold value larger than the second threshold during the execution of the threatening operation. The security device according to claim 6, wherein the security device is executed.   The security action control means switches the security action to be executed from the threat action to the warning action when a sensor installed near the opening of the building stops reacting during the threat action. The security device according to claim 4, wherein   During the execution of at least one of the plurality of security operations, the security operation control means continues for a certain period of time when the certainty level corresponding to a human is lower than the certainty level corresponding to an object category other than a human being The security device according to claim 1, wherein all security operations are terminated when it is performed.   In addition to outputting the warning sound from the sound output device, the warning operation further activates a limited notification device that notifies the site that an abnormality has occurred in the site. The security device according to claim 4.   In addition to outputting the threatening sound from the sound output device, the threatening operation further activates a surrounding notification device that notifies the inside and outside of the site that an abnormality has occurred in the site. The security device according to claim 4, wherein: A security operation switching method that is executed in a security device that executes at least one of a plurality of predetermined security operations when an abnormality is detected in a monitoring area,
Obtaining a sensor response signal taking a binary value as to whether or not the sensor is reacting from a plurality of sensors installed in the monitoring area ;
Based on the time-series data of the sensor response signal, for each of the plurality of objects categories, the steps of the object belonging to each object category to compute confidence factor representing the likelihood that a sensor reaction factor,
Before xenon capacitors reaction signal, on the basis of said confidence, security operation switching method characterized by comprising the steps of: switching the security operation to be executed.

Images