JP7339755B2 - Abnormal sound monitoring system and abnormal sound monitoring method - Google Patents

Description

The present disclosure relates to an abnormal sound monitoring system and an abnormal sound monitoring method.

Conventionally, there is known a technique for estimating the position and type of an abnormal sound source, photographing it with a pan-tilt-zoom camera, and illuminating it with lighting to facilitate identification of the cause of the sound.

JP 2017-28529 A

However, in order to detect the situation inside the building in detail, it was necessary to install various sensors and cameras in many places.

An object of the present disclosure is to provide an abnormal sound monitoring system and an abnormal sound monitoring method capable of accurately and easily detecting an abnormal state in a building.

An abnormal sound monitoring system according to one aspect of an embodiment of the present invention includes a sound collector that collects sounds in a building, and an abnormal sound identifier that extracts abnormal sounds from sound information collected by the sound collector. and sound source localization for estimating the position of the sound source of the sound information collected by the sound collecting unit, and estimating the number and moving direction of the persons emitting the abnormal sound if the abnormal sound is caused by a human being. , information on the abnormal sound extracted by the abnormal sound identification unit, and information on the position of the sound source, the number of people, and the direction of movement estimated by the sound source localization unit. an abnormal state determination unit for calculating a risk value indicating the abnormal state determination unit, the abnormal state determination unit obtains a basic risk value based on the information on the abnormal sound extracted by the abnormal sound identification unit, and estimates the basic risk value by the sound source localization unit. A correction value is obtained based on the position of the sound source of the abnormal sound, the number of people, or the direction of movement, and the risk value of the abnormal sound is calculated by multiplying the basic risk value by the correction value.

Similarly, an abnormal sound monitoring method according to an aspect of an embodiment of the present invention includes a sound collecting step of collecting sound in a building, and extracting an abnormal sound from the sound information collected in the sound collecting step. an abnormal sound identifying step; and estimating the sound source position of the sound information collected in the sound collecting step, and if the abnormal sound is generated by a human, the number and movement direction of the human beings emitting the abnormal sound. on the basis of the sound source localization step of estimating , the information of the abnormal sound extracted in the abnormal sound identification step, and the information of the sound source position, the number of people, and the movement direction estimated in the sound source localization step, and an abnormal state determination step of calculating a risk value indicating the degree of risk of the abnormal sound, wherein the abnormal state determination step calculates a basic risk value based on the information of the abnormal sound extracted in the abnormal sound identification step. a correction value is obtained based on the sound source position of the abnormal sound estimated in the sound source localization step, the number of people, or the movement direction , and the abnormal sound is obtained by multiplying the basic risk value by the correction value. to calculate the risk value of

Advantageous Effects of Invention According to the present disclosure, it is possible to provide an abnormal sound monitoring system and an abnormal sound monitoring method capable of accurately and easily detecting an abnormal state in a building.

BRIEF DESCRIPTION OF THE DRAWINGS It is a figure explaining the outline|summary of the abnormal sound monitoring system which concerns on embodiment. 1 is a functional block diagram of an abnormal sound monitoring system according to an embodiment; FIG. It is a figure which shows an example of a structure of a basic risk value calculation table. It is a figure which shows an example of a structure of a correction value calculation table. It is a figure which shows an example of an abnormal sound alerting|reporting screen. 4 is a flowchart of abnormal sound monitoring processing performed by the abnormal sound monitoring system; FIG. 5 is a schematic diagram showing an example of an abnormal sound generation case; FIG. 8 is a diagram showing a list of risk values calculated by the abnormal sound monitoring system according to the abnormal sound generation cases of FIG. 7; FIG. 9 is a graph showing changes in risk values according to the chronological order of FIG. 8; FIG. FIG. 9 is a graph showing changes in risk value estimation according to the chronological order of FIG. 8; FIG.

Embodiments will be described below with reference to the accompanying drawings. In order to facilitate understanding of the description, the same constituent elements in each drawing are denoted by the same reference numerals as much as possible, and overlapping descriptions are omitted.

FIG. 1 is a diagram illustrating an outline of an abnormal sound monitoring system 1 according to an embodiment. As shown in FIG. 1, the abnormal sound monitoring system 1 according to this embodiment is applied to a building 100 in which the state of the room is monitored by a microphone 21 and a camera 22 .

As shown in FIG. 1 , the abnormal sound monitoring system 1 detects the occurrence of an abnormal state inside the building 100 based on sound information from the microphone 21 . In the example of FIG. 1, a case in which a thief breaks into a building 100, destroys a safe, and steals valuables inside is illustrated as an example of an abnormal state. In this case, the abnormal sound monitoring system 1 generates a plurality of abnormal sounds in time series, such as the sound of breaking glass near the window, the sound of a person entering through the window, the sound of footsteps walking in the room, the noise of destroying a safe, and the like. , and calculate the degree of risk based on the combination of such abnormal sounds.

Further, when an abnormality is detected, the abnormal sound monitoring system 1 estimates the position of the sound source of the abnormal sound, photographs the sound source position with the camera 22, and acquires image data of the sound source position of the abnormal sound. When the abnormal sound monitoring system 1 detects an abnormal state, the information is transmitted to the information processing terminal 23 carried by the security guard, and the information is displayed on the screen to notify the occurrence of the abnormality. At this time, the camera image is also output to the information processing terminal 23 together.

FIG. 2 is a functional block diagram of the abnormal sound monitoring system 1 according to the embodiment. As shown in FIG. 1, the abnormal sound monitoring system 1 includes a sound collector 2, an abnormal sound identification unit 3, a sound source localization unit 4, an abnormal state determination unit 5, and a camera control unit 6 (imaging control unit). , and a screen generator 7 .

The sound collector 2 collects sounds inside the building 100 . Specifically, the sound collector 2 collects a plurality of pieces of sound information detected by the microphones 21 installed inside the building 100 . The microphone 21 may be, for example, a plurality of monaural microphones installed inside the building 100, or may be a multi-array microphone.

The abnormal sound identifying section 3 extracts abnormal sounds from the sound information collected by the sound collecting section 2 . The abnormal sound identification unit 3 has, for example, information on samples of a plurality of types of abnormal sounds, and calculates the degree of similarity with the sample from the frequency feature amount of the sound data collected by the sound collection unit 2 to identify which abnormal sound. Sound or identification can be performed. It should be noted that the technology disclosed in Japanese Patent No. 5767825 can be applied to the method of identifying everyday sounds/abnormal sounds in the abnormal sound identification unit 3 . When the abnormal sound identification unit 3 identifies an abnormal sound, the abnormal sound identification unit 3 outputs information such as the type of abnormal sound, the tool generating the abnormal sound, and the time of occurrence to the abnormal state determination unit 5 .

The sound source localization unit 4 estimates the sound source position of the sound information collected by the sound collection unit 2 . The sound source localization unit 4 can estimate the sound source position of the sound information based on information such as volume and generation time of each of the sound information in the building 100 collected by the sound collection unit 2, for example.

Also, the sound source localization unit 4 can estimate the number of people who are emitting abnormal sounds based on a plurality of sound information collected by the sound collection unit 2 . For example, when the sound information can be determined as footsteps, the number of people can be estimated by calculating the number of pedestrians. Furthermore, the sound source localization unit 4 can detect the movement direction of the pedestrian, who is the sound source, by using the time difference and amplitude difference of the sound information collected by the sound collection unit 2 . The technique disclosed in Japanese Patent No. 5721470 can be applied to the method of estimating the number of people and the direction of movement in the sound source localization unit 4 .

The sound source localization unit 4 outputs information such as the position of the sound source of the abnormal sound detected by the abnormal sound identification unit 3, the number of people, and the direction of movement to the abnormal state determination unit 5. FIG.

The abnormal state determination unit 5 determines what is happening in the building from the results of abnormal sound identification by the abnormal sound identification unit 3 and the sound source localization unit 4 and the time transition. For example, based on abnormal sound time-series information including multiple abnormal sounds, it is possible to determine whether a series of abnormal events including multiple abnormal sounds, for example, a suspicious person such as a thief, has entered the building. In this case, the abnormal state determination unit 5 determines the state of intrusion, and outputs the presence or absence of intruders, the number of intruders, tools, and the degree of danger. The number of people is determined using the amount of footsteps and abnormal sounds detected at the same time. Tools (belongings) are estimated from the type of breaking sound. Regarding the degree of danger, comprehensively judge from the number of people, tools, and escape situation.

For example, the time-series information of the abnormal sound and the abnormal state determination result of the abnormal state determination unit 5 based on this time-series information include the following patterns.

<Example 1>
・Abnormal sound time-series information: Glass breaking sound (breaking) ⇒ Window opening and closing sound ⇒ Landing sound (multiple) ⇒ Walking sound (multiple) ⇒ Safe breaking sound (prying open) ⇒ Walking sound (multiple) ⇒ Door opening and closing sound ・Abnormal state judgment Result: burglary and theft by multiple people, escape aiming at the safe with a crowbar, medium risk

<Example 2>
・Abnormal sound time-series information: Glass breaking sound (prying) ⇒ Window opening/closing sound ⇒ Landing sound (singular) ⇒ Walking sound (singular) ⇒ Object sound (opening/closing sound) ⇒ Walking sound (singular) ⇒ Object sound (opening/closing sound)
・ Abnormal state judgment result: Intrusion by singular, possessing a driver, searching, high risk

<Example 3>
・Abnormal sound time-series information: door unlock sound ⇒ walking sound (singular) ⇒ object sound ⇒ walking sound (singular) ⇒ voice (singular)
・ Abnormal state judgment result: Intrusion by singular or mistaken entry during security, possession of key, danger level medium

<Example 4>
The situation is judged when the security guard checks the interior, and if the security guard encounters a thief and falls into a dangerous situation, the content is output.
・Abnormal sound time-series information: door unlocking sound ⇒ walking sound (singular) ⇒ voice (scream) ⇒ voice (fighting) ⇒ hitting sound

In this embodiment, in order to realize the above functions, the abnormal state determination unit 5 combines the abnormal sound information extracted by the abnormal sound identification unit 3 with the sound source position information estimated by the sound source localization unit 4. Based on this, a risk value indicating the degree of risk of a series of abnormal events including multiple abnormal sounds is calculated. More specifically, the abnormal state determination unit 5 first calculates a risk value indicating the degree of risk of each abnormal sound based on the abnormal sound information and the sound source position information. Then, a plurality of calculated risk values are integrated to calculate a time transition of the risk values relating to a series of abnormal events including a plurality of abnormal sounds.

The abnormal state determination unit 5 has a basic risk value calculation table 8, a correction value calculation table 9, and a risk value calculation unit 10 as elements for realizing the above functions.

The basic risk value calculation table 8 stores the type of abnormal sound and the basic risk value in association with each other. FIG. 3 is a diagram showing an example of the configuration of the basic risk value calculation table 8. As shown in FIG. As shown in FIG. 3, the basic risk value calculation table 8 stores, for example, the type of abnormal sound, the details of the abnormal sound, the tool used for the abnormal sound, and the information of each item of the basic risk value in association with each other. are doing. In the present embodiment, the risk value set for each abnormal sound alone is called a "basic risk value" in order to distinguish it from the risk value finally calculated by the abnormal state determination unit 5. FIG.

In the basic risk value calculation table 8, an individual basic risk value is assigned to each of a plurality of types of abnormal sounds. For example, a relatively large basic risk value is set for an abnormal sound such as glass breaking sound, metal breaking sound, drilling sound, scream, etc., which is considered to be of high danger if a guard were to face it. On the other hand, a relatively small basic risk value is set for an abnormal sound such as a door operation sound or a window operation sound, which is considered not to pose a high degree of danger even if a guard were to face it. Further, when there is no sound, the basic risk value is set to 0, since it is assumed that the situation in the building 100 has not changed.

In addition, in the basic risk value calculation table 8, different basic risk values are set according to the details of the abnormal sound and the tool assumed to be possessed even for the same type of abnormal sound. For example, if the type of abnormal sound is the sound of breaking glass, the basic risk value is set to a relatively large value for the sound of breaking with a crowbar or a brick, and the relative risk value is set for the sound of breaking with a screwdriver. is set to a small value. When the type of abnormal sound is voice, the basic risk value is set to a relatively large value for screams and fighting voices, and is set to a relatively small value for normal conversational voices.

The basic risk value calculation table 8 is used, for example, by the risk value calculator 10 to extract the basic risk value associated with the type of abnormal sound, its details, and tool information. be. Information estimated by the abnormal sound identification unit 3 can be used as the type and details of the abnormal sound and information on the tool.

The correction value calculation table 9 stores various abnormal sound conditions and correction values in association with each other. FIG. 4 is a diagram showing an example of the configuration of the correction value calculation table 9. As shown in FIG. As shown in FIG. 4, the correction value calculation table 9 stores, for example, various conditions of abnormal sounds, details of the conditions, and information on each item of weight (correction value) in association with each other.

In the correction value calculation table 9, an individual correction value is assigned to each of multiple types of conditions. Also, even under the same conditions, different correction values are set according to the details of the conditions. For example, if the condition is the number of people, the correction value is set to 1 if there is only one person, and the correction value is set to increase as the number of people increases. Further, when the condition is the time of occurrence, different correction values are set depending on the time period. For example, a relatively high correction value is set for the midnight time period.

The risk value calculator 10 calculates the risk value of each abnormal sound by multiplying the basic risk value extracted from the basic risk value calculation table 8 and the correction value extracted from the correction value calculation table 9 . Further, by integrating the plurality of calculated risk values, the temporal transition of the risk values relating to a series of abnormal events including a plurality of abnormal sounds is calculated.

The camera control unit 6 is an element that controls the operation of the camera 22 (imaging unit), and causes the camera 22 to image the position where the abnormal sound is generated. As the camera 22, for example, a pan-tilt-zoom camera that can be remotely controlled can be used.

The screen generation unit 7 generates an abnormal sound notification screen 11 including information regarding the risk value calculated by the abnormal state determination unit 5 . The screen generator 7 can output the generated abnormal sound notification screen 11 to an external device such as the information processing terminal 23 carried by the observer to inform the observer of the occurrence of an abnormal event.

FIG. 5 is a diagram showing an example of the abnormal sound notification screen 11. As shown in FIG. As shown in FIG. 5, the abnormal sound notification screen 11 includes, for example, a map display unit 12 that superimposes the occurrence positions and movement transitions of a plurality of abnormal sounds on a three-dimensional map of the monitoring space of the building 100, and each abnormal sound. and a list display section 13 listing various types of information for each. The abnormal sound notification screen 11 may include only one of the map display section 12 and the list display section 13 . Alternatively, the abnormal sound notification screen 11 may display only the map display section 12 as a basic screen, and when an abnormal sound object in the map display section 12 is clicked, information in the list display section 13 may be popped up. .

The screen generator 7 can also include the camera image of the abnormal sound generation position captured by the camera 22 by the camera controller 6 in the abnormal sound notification screen 11 .

In this way, the screen generation unit 7 can visualize the abnormal state by fitting the obtained abnormal state and transition of the abnormal sound source to the drawing of the building 100 . The screen generation unit 7 transmits information (for example, an HTML file) necessary for screen generation to the information processing terminal 23, and uses the information received by the information processing terminal 23 to display the abnormal sound notification screen 11 on the display screen. It may be configured to be displayed.

The abnormal sound monitoring system 1 physically includes a CPU (Central Processing Unit), RAM (Random Access Memory) and ROM (Read Only Memory) as main storage devices, a communication module as a data transmission/reception device, an auxiliary storage device, It can be configured as a computer device including, for example, a circuit board. Each function of the abnormal sound monitoring system 1 described above operates a communication module or the like under the control of the CPU by loading predetermined computer software onto hardware such as the CPU and RAM, and operates the RAM and auxiliary storage. It is realized by reading and writing data in the device.

Next, an abnormal sound monitoring method according to this embodiment will be described with reference to FIG. FIG. 6 is a flowchart of abnormal sound monitoring processing performed by the abnormal sound monitoring system 1. FIG.

In step S01, the sound in the building 100 is collected by the sound collector 2 (sound collection step).

In step S02, the abnormal sound identification unit 3 extracts an abnormal sound from the sound information collected in step S01 (abnormal sound identification step), and determines whether or not an abnormal sound is detected. If an abnormal sound is detected (Yes in step S02), the process proceeds to step S03, and if no abnormal sound is detected (No in step S02), the process returns to step S01. The abnormal sound identification unit 3 outputs information on the abnormal sound detected in step S<b>02 to the sound source localization unit 4 and the abnormal state determination unit 5 .

In step S03, the sound source localization unit 4 estimates the sound source position of the abnormal sound detected in step S02 (sound source localization step).

In step S04, when the abnormal sound detected in step S02 is footsteps, the sound source localization unit 4 estimates the number of people involved in the abnormal sound.

In step S05, if the abnormal sound detected in step S02 is footsteps, the sound source localization unit 4 estimates the direction of the intrusion of the abnormal sound. The sound source localization unit 4 outputs each information estimated in steps S03 to S05 to the abnormal state determination unit 5. FIG.

In step S06, the risk value calculator 10 of the abnormal state determination unit 5 refers to the basic risk value calculation table 8 and extracts a basic risk value based on the type of abnormal sound.

In step S07, the risk value calculation unit 10 of the abnormal state determination unit 5 extracts a correction value by referring to the correction value calculation table 9 based on the abnormal sound generation condition.

In step S08, the risk value calculation unit 10 of the abnormal state determination unit 5 multiplies the basic risk value extracted in step S06 by the correction value extracted in step S07 to calculate the risk value of the abnormal sound. (abnormal state determination step).

In step S09, the screen generation unit 7 generates the abnormal sound notification screen 11 (see FIG. 5) including information regarding the risk value calculated in step S08. The screen generator 7 transmits the generated abnormal sound notification screen 11 to an external device such as the information processing terminal 23 carried by the surveillance staff.

In step S10, it is checked whether the abnormal sound monitoring system 1 is still operating. If it is in operation (Yes in step S10), the process returns to step S01 to continue the abnormal sound monitoring process. When the operation of the system ends (No in step S10), this control flow ends.

Calculation examples of the risk value by the abnormal sound monitoring system 1 will be described with reference to FIGS. 7 to 10. FIG. FIG. 7 is a schematic diagram showing an example of an abnormal sound generation case. As shown in FIG. 7, the site image in this example is a building in which a store and an office are connected by a door, and a safe is inside the office. In addition, the story in this example is that three thieves broke the window glass of the store (1 in the figure), entered the store (2 in the figure), and opened the door (3 in the figure). The flow is to break into the office (4 in the figure), destroy the safe in the office (5 in the figure), take out the contents of the safe and escape (6 and 7 in the figure).

FIG. 8 is a diagram showing a list of risk values calculated by the abnormal sound monitoring system 1 according to the abnormal sound generation cases of FIG. Fig. 8 shows the type of abnormal sound, the basic risk value of each abnormal sound, the weight correction value according to various conditions of each abnormal sound (time of occurrence, place of occurrence, number of people, direction of approach), and each abnormality after weighting calculation. The danger value of the sound and the accumulated danger value are shown for each abnormal sound in chronological order.

FIG. 9 is a graph showing changes in risk values according to the chronological order of FIG. The horizontal axis of FIG. 9 indicates the time transition according to the time series order of FIG. The vertical axis in FIG. 9 indicates the risk value. As shown in the "Risk Value" column in Fig. 8 and in Fig. 9, during the period when the thief invades the room at the beginning of the story and moves to the safe, the danger value changes at a certain level, and the danger increases as the safe approaches. The value rises, and the danger value is maximum when the safe is destroyed. After that, when the thief starts to run away, the danger value changes to a negative value.

FIG. 10 is a graph showing changes in risk value estimation according to the chronological order of FIG. The horizontal axis of FIG. 10 indicates the time transition according to the time series order of FIG. The vertical axis in FIG. 10 indicates the cumulative risk value. As shown in the "total danger value" column in FIG. 8 and in FIG. , it decreases when the thief starts to run away, and becomes constant when the thief runs away from the building and becomes silent.

Next, the effects of the abnormal sound monitoring system 1 will be described. The abnormal sound monitoring system 1 of this embodiment includes a sound collector 2 that collects sounds in the building, an abnormal sound identifier 3 that extracts abnormal sounds from the sound information collected by the sound collector 2, A sound source localization unit 4 for estimating the sound source position of sound information collected by the sound unit 2, abnormal sound information extracted by the abnormal sound identification unit 3, and sound source position information estimated by the sound source localization unit 4. and an abnormal state determination unit 5 for calculating a risk value indicating the degree of risk of abnormal sound based on the above.

With this configuration, it is possible to detect an abnormal sound and estimate the degree of risk associated with the abnormal sound using only the sound information collected by the sound collector 2. Good and easy to detect.

Further, in the abnormal sound monitoring system 1 of the present embodiment, the abnormal state determination unit 5 includes a basic risk value calculation table 8 that stores abnormal sound types and basic risk values in association with each other, various conditions of abnormal sounds, and correction values. and the basic risk value extracted from the basic risk value calculation table 8 and the correction value extracted from the correction value calculation table 9 are multiplied to obtain the risk value of each abnormal sound. and a risk value calculation unit 10 for calculating the

With this configuration, by referring to the basic risk value calculation table 8 and the correction value calculation table 9 based on information such as the type of abnormal sound and the position of the sound source, the basic risk value of the abnormal sound is calculated in consideration of various factors. It can be calculated with high accuracy, and the basic risk value can be appropriately weighted and corrected to a more appropriate value, so that the degree of risk can be estimated with even higher accuracy. Further, since the risk value can be calculated by a simple calculation of extracting the basic risk value and the correction value by referring to the basic risk value calculation table 8 and the correction value calculation table 9 and multiplying the two, the risk value can be calculated more easily. state can be detected.

In addition, in the abnormal sound monitoring system 1 of the present embodiment, as described with reference to FIG. Calculate the time transition of the risk value related to the abnormal event. With this configuration, it is possible to estimate not only the degree of risk at each point in the time-series information of abnormal sounds, but also the degree of risk at present based on a series of abnormal sounds. Such current risk information can be used as an index for comprehensively judging the risk of a series of abnormal events including a plurality of abnormal sounds, and is useful for the judgment of the surveillance personnel of the building 100 .

The abnormal sound monitoring system 1 of the present embodiment also includes a screen generation unit 7 that generates the abnormal sound notification screen 11 including information on the risk value calculated by the abnormal state determination unit 5 . With this configuration, it is possible to visualize information related to the occurrence of an abnormal event in the building 100 and provide it to the surveillance staff of the building 100 in an easy-to-understand manner. Providing more detail on the abnormal conditions within the building 100 allows the lifeguards to respond more quickly.

In addition, the abnormal sound monitoring system 1 of the present embodiment includes a camera control section 6 that causes the camera 22 that captures an image of the inside of the building to capture an image of the position where the abnormal sound is generated. With this configuration, when an abnormal state occurs, a camera image of the position where the abnormal sound is generated can be obtained, and if the obtained camera image is provided to the monitor along with the abnormal sound notification screen 11, the information regarding the occurrence of the abnormal event can be more easily understood. can be provided, thereby allowing the observer to make a more accurate judgment on abnormal conditions.

The present embodiment has been described above with reference to specific examples. However, the present disclosure is not limited to these specific examples. Design modifications to these specific examples by those skilled in the art are also included in the scope of the present disclosure as long as they have the features of the present disclosure. Each element included in each specific example described above and its arrangement, conditions, shape, etc. are not limited to those illustrated and can be changed as appropriate. As long as there is no technical contradiction, the combination of the elements included in the specific examples described above can be changed as appropriate.

1 Abnormal sound monitoring system 2 Sound collector 3 Abnormal sound identification unit 4 Sound source localization unit 5 Abnormal state determination unit 6 Camera control unit (imaging control unit)
7 screen generation unit 8 basic risk value calculation table 9 correction value calculation table 10 risk value calculation unit 21 microphone 22 camera (imaging unit)
23 information processing terminal 100 building

Claims (6)

a sound collecting unit that collects sound in the building;
an abnormal sound identification unit that extracts an abnormal sound from the sound information collected by the sound collecting unit;
a sound source localization unit for estimating the sound source position of the sound information collected by the sound collecting unit, and estimating the number of persons and moving directions of the persons emitting the abnormal sound if the abnormal sound is caused by a person; ,
A risk value indicating the degree of risk of the abnormal sound based on the information about the abnormal sound extracted by the abnormal sound identification unit and the information about the sound source position, the number of people, and the direction of movement estimated by the sound source localization unit. an abnormal state determination unit that calculates
with
The abnormal state determination unit
obtaining a basic risk value based on the information of the abnormal sound extracted by the abnormal sound identification unit;
obtaining a correction value based on the sound source position of the abnormal sound estimated by the sound source localization unit, the number of people, or the moving direction ;
calculating the risk value of the abnormal sound by multiplying the basic risk value and the correction value;
Abnormal sound monitoring system. The abnormal state determination unit
a basic risk value calculation table that associates and stores information on abnormal sounds including types of abnormal sounds and basic risk values;
a correction value calculation table that associates and stores correction values with various conditions of the abnormal sound, including the position of the sound source of the abnormal sound, the number of people, and the direction of movement;
The basic risk value extracted from the basic risk value calculation table based on the information of the abnormal sound extracted by the abnormal sound identification unit, and the sound source position, the number of people, or the movement estimated by the sound source localization unit. a risk value calculation unit that calculates the risk value of each abnormal sound by multiplying the correction value extracted from the correction value calculation table based on the direction;
having
The abnormal sound monitoring system according to claim 1. The abnormal state determination unit integrates the calculated plurality of risk values to calculate a time transition of the risk values related to a series of abnormal events including a plurality of the abnormal sounds.
The abnormal sound monitoring system according to claim 1 or 2. a screen generation unit that generates an abnormal sound notification screen including information about the risk value calculated by the abnormal state determination unit;
The abnormal sound monitoring system according to any one of claims 1 to 3. An imaging control unit for imaging the position where the abnormal sound is generated is provided in the imaging unit for imaging the inside of the building,
The abnormal sound monitoring system according to any one of claims 1 to 4. a sound collecting step for collecting sound in the building;
an abnormal sound identifying step of extracting an abnormal sound from the sound information collected in the sound collecting step;
A sound source localization step of estimating the position of a sound source of the sound information collected in the sound collecting step, and estimating the number and movement direction of the persons emitting the abnormal sound if the abnormal sound is caused by a human being. and,
The degree of risk of the abnormal sound is indicated based on the information on the abnormal sound extracted in the step of identifying the abnormal sound and the information on the position of the sound source, the number of people, and the direction of movement estimated in the step of localizing the sound source. an abnormal condition determination step of calculating a risk value;
including
The abnormal state determination step includes:
obtaining a basic risk value based on the abnormal sound information extracted in the abnormal sound identification step;
obtaining a correction value based on the sound source position, the number of people, or the moving direction of the abnormal sound estimated in the sound source localization step;
calculating the risk value of the abnormal sound by multiplying the basic risk value and the correction value;
Abnormal sound monitoring method.

Images