JP7103901B2 - Security system and security management system - Google Patents

Description

The present invention relates to a crime prevention system and a crime prevention management system.

In buildings such as houses, illegal invasion through openings such as doors and windows is wary, and a security system that enhances the security performance of the building by photographing the vicinity of the openings from the outdoor side has been proposed. In such a crime prevention system, it is desirable to detect a suspicious person who is trying to trespass as soon as possible, and therefore, various methods for automatically distinguishing a suspicious person from a person other than the suspicious person based on the captured image. The system has been proposed. For example, Patent Document 1 proposes to detect a fraudulent act (suspicious person) by determining a person's movement such as a height of a posture, a moving speed, and a moving direction based on a captured image. There is.

Japanese Unexamined Patent Publication No. 2012-104022

When trying to detect a suspicious person based on the movement of a person on the outdoor side of a building as in the technique of Patent Document 1, the movement of the suspicious person and a person other than the suspicious person (for example, a resident or a visit other than the resident) It is necessary to identify the behavior of the person, etc.). However, for example, the action of a suspicious person who attempts to break into the building by destroying the locking device provided on the front door is similar to the action of a resident who has returned home to unlock the front door. These are difficult to identify.

Therefore, if an attempt is made to detect a suspicious person on the outdoor side of a building based only on the movement of a person, a person other than the suspicious person such as a resident may be erroneously detected as a suspicious person.

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a crime prevention system and a crime prevention management system that quickly and accurately estimate a suspicious person.

In order to solve the above problems, the security system of the first invention is a security system used for crime prevention of a structure including an opening / closing body for opening / closing an opening and a locking device for locking the opening / closing body.
A photographing means for photographing a predetermined range in the vicinity of the opening / closing body as a photographing area, and
An extraction means for extracting an object from an image captured by the photographing means,
An motion information collecting means that tracks the motion of an object extracted by the extraction means and collects motion information of the object, and an motion information collecting means.
A state estimating means for estimating whether or not the motion information of the object collected by the motion information collecting means is in a predetermined abnormal state, and
It is characterized by having.

The structure can be applied to various structures such as commercial buildings, warehouses, vehicles, and safes in addition to residential buildings, but for convenience of explanation, the following is a typical example of a residential building. The operation and effect of each invention will be described. In the first invention, for example, a suspicious person invades a building by destroying a door or a window provided at an opening of the building, or a locking device or the like (hereinafter, an opening / closing body or the like) provided therein. try to. In this case, it is considered that these opening / closing bodies and the like are destroyed by using some kind of tool (hereinafter referred to as a destructive tool). The destructive tool applies a destructive force to the opening / closing body and the like. In other words, the opening / closing body or the like is destroyed by using the destructive tool with a characteristic operation with respect to the opening / closing body or the like. According to the first invention, the destructive action on the opening / closing body or the like is executed by paying attention to the characteristic movement of the opening / closing body or the like by such a destructive tool and specifying the movement of the object on the opening / closing body or the like. Estimate whether or not.

That is, according to the first invention, an object is extracted from the image captured by the photographing means, and it is determined whether or not it is in an abnormal state based on the operation of the object. When the resident who has returned home performs a normal unlocking operation, the movement of the object, which is considered to be a destructive act on the opening / closing body, is not detected. In addition, even for a visitor such as a courier, there may be various situations such as the presence or absence of luggage, but the movement of an object that is considered to be a destructive act on the opening / closing body or the like is not detected. Therefore, a person other than the suspicious person is erroneously detected as a suspicious person by determining whether or not a person near the opening is a suspicious person trying to invade the building based on the movement of the object which is considered to be a vandalism. You can avoid that.

Further, according to the first invention, since the suspicious person is detected based on the movement of the object performed at the time of destruction, it is possible to detect the abnormal state before the suspicious person invades the building. As a result, it is possible to respond more quickly to the intrusion of a suspicious person, and the damage can be minimized.

The security system of the second invention is the security system of the first invention.
The extraction means is characterized in that an image of a person is extracted from an image captured by the photographing means, and then an object possessed by the person is extracted.

For example, when an object is left near the opening, the object may move due to wind or the like. In such a case, for example, if the motion of the abandoned object is determined to be a destructive action, it will be erroneously detected as an abnormal state. According to the second invention, first, a person is recognized in the captured image, and the object possessed by the recognized person is focused on. Thereby, the movement of the object caused by the movement performed by the person using the object can be specified. That is, since the destructive action on the opening / closing body or the like is performed by the person and the object, the destructive action can be detected more reliably by estimating the abnormal state based on the movement of the object possessed by the person, and the false detection can be reduced. ..

The security system of the third invention is the security system of the first or second invention.
An object identifying means for identifying that the object extracted by the extracting means is a predetermined object requiring attention is provided.
The abnormal state is characterized in that the object requiring attention performs a predetermined operation.

According to the third invention, an object requiring attention and a specific action for the object requiring attention are predetermined. For example, a person other than a suspicious person such as a resident may possess or use an object requiring attention. In such a case, even if the object requiring attention is possessed, the movement of the object, which is considered to be a destructive act on the opening / closing body or the like using the object requiring attention, is not detected. According to the third invention, since it is determined whether or not an abnormal state is present based on the object requiring attention and its operation, erroneous detection can be reduced and the accuracy of detecting the abnormal state can be improved.

The crime prevention system of the fourth invention is characterized in that, in the third invention, the object requiring attention is an object having a long shape and a flat surface at the tip.

According to the fourth invention, the shape of the object requiring attention is defined. For example, in the case of unauthorized intrusion through the opening, a breaking tool is inserted into the gap between the opening and the opening / closing body, and a force is applied to the insertion portion of the breaking tool to provide the opening and the opening / closing body. In many cases, the engagement structure with the locking device (locked state of the locking device) is destroyed. Therefore, the destructive tool is often an object that is long and has a flat surface at the tip because of its ease of insertion into a gap, ease of applying a force for destruction, and the like. According to the fourth invention, by determining the shape of the object requiring attention, the object requiring attention can be detected according to the shape of the object based on the image captured by the photographing means, so that the accuracy of detecting the abnormal state is further improved. be able to.

The security system of the fifth invention is the fourth invention.
The state estimation means is in an abnormal state when the motion information collecting means collects motions of reciprocating and reciprocating within a range of a predetermined angle a predetermined number of times with the tip of the object requiring attention as a fulcrum. It is characterized by presuming that.

According to the fifth invention, the specific operation of the object requiring attention when the opening / closing body or the like is destroyed by using the object requiring attention is defined. A suspicious person inserts a flat surface, which is the tip (one end side) of the object requiring attention, into the gap between the opening and the opening / closing body, uses the tip as a fulcrum, and applies the principle of the lever from the other end side of the object requiring attention. Use it to apply force to the tip. Specifically, the tip portion of the object requiring attention is used as a fulcrum, and the other end side of the object requiring attention is pushed and pulled within a predetermined movement range (the action of applying a force toward the fulcrum) to perform the tip portion. The engagement structure between the opening and the locking device provided on the opening / closing body (locked state of the locking device) is often destroyed. That is, according to the fifth invention, the abnormal state is determined based on the vandalism using the object requiring attention when a suspicious person illegally invades through the opening. Therefore, since the abnormal state is determined based on the specific vandalism, the accuracy of detecting the abnormal state can be further improved.

The security system of the sixth invention is defined in any one of the first to fifth inventions.
The state estimation means is characterized by comprising a determination standard generating means for generating a determination criterion of an abnormal state by executing deep learning using the information collected by the operation information collecting means.

According to the sixth invention, since the criterion for determining the abnormal state is generated by deep learning based on the information collected by the motion information collecting means, it is possible to improve the specific accuracy of the motion of the object in the abnormal state. It can reduce false positives.

The security management system of the seventh invention is
The security system described in any of the first to sixth is provided.
The photographing means is provided corresponding to each of the plurality of opening / closing bodies.
It is equipped with a management center that manages the security system.
When the state estimation means presumes that the predetermined opening / closing body is in an abnormal state, the security system notifies the management center of the warning information indicating the abnormal state by associating it with the predetermined opening / closing body. It is characterized by doing.

According to the seventh invention, when it is presumed to be in an abnormal state, warning information indicating the abnormal state is notified to the management center. The management center can promptly take measures such as going to the building and checking the safety inside and outside the building by being notified of the warning information. As a result, it is possible to catch a suspicious person or narrow the time for the suspicious person who has invaded the building, and the damage can be minimized.

The plan view which shows the structure around a building in this embodiment. The figure which shows the electrical composition of a security system. A conceptual diagram showing the stored contents of the first area provided in the object database of the controller to be watched. A flowchart showing a processing procedure of security processing. An example of an image taken by a camera, FIG. 5A is an image showing a series of operations in which a resident unlocks the entrance door, and FIG. 5B is provided on the entrance door by a suspicious person. An image showing a series of operations in which a locked locking device is destroyed.

Hereinafter, an embodiment embodying the present invention will be described with reference to the drawings. FIG. 1 is a plan view showing the area around the entrance of the building on the site.

In FIG. 1, a building 10 is provided in the site 11. The building 10 is provided with an entrance 12 as an entrance / exit. It is possible to enter and exit the building 10 through the entrance 12, and the entrance 12 is provided with an entrance door 13 as an opening / closing body. An entrance porch 15 is provided between the entrance 12 of the building 10 and the public road 14. On the outer edge of the site 11, an outer structure 16 such as a fence or a fence is provided at a portion other than the entrance porch 15.

Two locking devices 21 are attached to the entrance door 13 (indoor side) at an upper position and a lower position of the door. Note that, for convenience, only one locking device 21 is shown in FIG. The locking device 21 includes an electric lock that locks and unlocks by energization. The locking device 21 is locked and unlocked by an electronic key carried by a resident of the building 10.

As shown in FIG. 1, the locking device 21 has a dead bolt 22, and a receiving hole portion 23 corresponding to the dead bolt 22 is provided on the side surface of the entrance 12 (the side surface in the opening of the outer wall 17). ing. By causing the dead bolt 22 to appear and disappear by the locking device 21 and inserting or removing it into the receiving hole 23, the locking device 21 is locked or unlocked, and the opening and closing of the entrance door 13 is regulated or permitted. In other words, the dead bolt 22 is inserted into the receiving hole 23 to form an engaging structure, so that the locking device 21 is locked.

The entrance door 13 is provided with a door handle 18 as a handle on the outdoor side, and the door handle 18 has a built-in touch sensor (not shown). The touch sensor is a capacitance type sensor, and detects that a person comes into contact with the sensor portion. The locking device 21 is locked and unlocked by the touch sensor detecting the contact of a person while the electronic key carried by the resident is authenticated.

A camera 25 for photographing (imaging) a person around the entrance door 13 is provided near the entrance door 13 on the outdoor side. The camera 25 is oriented obliquely downward on the entrance door 13 side so as to photograph the vicinity of the entrance door 13 (particularly, the vicinity of the door handle 18). The photographing area SA by the camera 25 is set so that the whole body of a person who locks / unlocks the locking device 21 and opens / closes the entrance door 13 can be photographed around the entrance door 13. Specifically, at least on the outdoor side, it is possible to photograph a range of up to about 2 m upward from the floor surface near the entrance door 13.

Subsequently, the electrical configuration of the security system will be described with reference to FIG.

As shown in FIG. 2, the security system includes a controller 31 as a control means. The controller 31 includes a well-known microcomputer including a CPU, a ROM, a RAM, and the like, and is provided on the indoor side of the building 10 (for example, a wall surface around the entrance).

The controller 31 is provided with a database of objects requiring attention 32 for registering and managing various information relating to objects requiring attention, and an information storage area including a first area 32a and a second area 32b is set. The result of the abnormal state based on the motion information of the object requiring attention and the motion information of the object requiring attention collected by the camera 25 is registered in the second area 32b.

Here, the first area 32a will be described with reference to FIG. As shown in FIG. 3, the object information indicating the object requiring attention and the operation information of the object requiring attention are stored in the first area 32a. As shown in FIG. 3A, the object information of the object requiring attention is composed of identification information indicating the type of the object requiring attention and image information indicating an image of the object requiring attention. Bar is registered as an object to be watched. As shown in FIG. 3B, the motion information of the object requiring attention is registered as the motion information of the object requiring attention for estimating that it is in an abnormal state.

The controller 31 is electrically connected to the camera 25, and an image signal corresponding to the shooting content is output from the camera 25 to the controller 31. The controller 31 extracts (recognizes) a person around the entrance door 13 and an object possessed by the person based on the shooting result of the camera 25. The controller 31 recognizes, for example, a plurality of objects such as a bag and an umbrella possessed by a person.

As shown in FIG. 2, the controller 31 is provided with a timer unit 33 having a time measuring function, a clock function, and a calendar function. A communication unit 35 is connected to the controller 31, and the controller 31 can be connected to the network N by the communication unit 35. The communication unit 35 may be provided in the controller 31.

Here, the security system is introduced in a plurality of buildings 10A, 10B, ... 10N, and the security control center 50 centrally manages the resident data of the plurality of buildings 10A, 10B, ... 10N. A server 51 is installed in the security control center 50 and is managed by the administrator of the security control center 50.

The server 51 is mainly composed of a well-known microcomputer having a CPU or the like. The server 51 includes a control unit 52, a storage unit 53, and a communication unit 54. The communication unit 54 can communicate with an external device such as a controller 31 or a database of each building 10A, 10B, ... 10N via the network N. In this case, the control unit 52 can communicate with the external device (for example, the controller 31) via the communication unit 54. Then, various notification results transmitted from the controller 31 of each building 10A, 10B, ... 10N are stored in the storage unit 53.

Next, the warning process executed by the controller 31 will be described. FIG. 4 is a flowchart showing the process. This process is repeatedly executed in a predetermined time cycle (for example, 0.5 seconds). In the following, it is assumed that the building 10 has been registered as a user of the security control center 50.

First, in step S11, the camera 25 takes an image around the entrance door 13. In the following step S12, the image recognition process is executed. First, the image captured by the camera 25 is analyzed to extract feature quantities (color, shape, size, etc.), and the image of a person included in the captured image is recognized. Then, when the image of a person is recognized, the image of the person is analyzed to extract the feature amount (color, shape, size, etc.), and the image of the object included in the image of the person is recognized.

Here, a step of recognizing an object possessed by the user based on the image captured by the camera 25 using FIGS. 5A and 5B will be described. FIG. 5 (a) is an image showing a series of operations in which the entrance door 13 is unlocked by the resident, and FIG. 5 (b) shows a suspicious person destroying the locking device 21 provided on the entrance door 13. This is a video showing a series of operations. At this time, when the images of FIGS. 5A and 5B are acquired by the camera 25, the person is first recognized by the image recognition process by the controller 31, and the recognized object is possessed by the person. Is recognized (the recognized part is indicated by a broken line). Then, after performing the recognition processing of the person and the object in the image by the image recognition processing, the process proceeds to step S13.

In the following step S13, based on the result of the image recognition process in step S12, it is determined whether or not the image of a person is recognized, that is, whether or not there is a person around the entrance door 13. If the image of a person is not recognized, the alert process is terminated. On the other hand, when the image of a person is recognized, in step S14, it is determined whether or not the person in the vicinity of the entrance door 13 possesses the object. Here, in the image that has been image-recognized in step S12, it is determined whether or not the recognized person possesses an object. This process ends when the recognized person does not possess the object, that is, when the person near the entrance door 13 does not possess the object. If the recognized person possesses the object, that is, if the person near the entrance door 13 possesses the object, the process proceeds to step S15.

In step S15, by collating the recognized object possessed by the person (hereinafter referred to as possessed object) with the object information registered in the first area 32a, whether or not the possessed object corresponds to an object requiring attention. Is determined. If the possessed object does not correspond to the object requiring attention, this process ends. If the possessed object corresponds to an object requiring attention, the process proceeds to step S16.

In step S16, the operation of the possessed object (corresponding to the object requiring attention) is monitored. Specifically, the timer unit 33 measures the time, and the camera 25 captures an image of the vicinity of the entrance door 13 at a predetermined time. The predetermined time may be, for example, the time required for observing the movement of the object requiring attention presumed to be in an abnormal state, for example, 5 to 30 seconds, and 10 seconds in the present embodiment. After monitoring the movement of the possessed object with the camera 25 for 10 seconds, the process proceeds to step S17.

In the following step S17, the image taken by the camera 25 at a predetermined time is analyzed, and the motion information of the possessed object is collected based on the change in the predetermined time. Specifically, feature quantities (color, shape, size, etc.) are extracted for each captured image, and the image of a person included in the captured image and the object (possessed object) included in the image of that person. Recognize the image. Then, by analyzing the change in the recognized position of the possessed object, the motion of the possessed object is extracted, and the motion of the object at a predetermined time can be recognized.

For example, when the image (x) of FIG. 5 (b) is taken in step S11, the images (y) and (z) of FIG. 5 (b) are taken in step S16. Then, the possessed objects in the (y) and (z) images are recognized by the image recognition processing in the (y) and (z) images of FIG. 5 (b), and the (x), (y) and (z) images are converted into the possessed objects. Based on this, the movement of the possessed object can be grasped. Specifically, in the present embodiment, the operation of the possessed object can be grasped by the difference between the position of the possessed object in the (y) image and the position of the possessed object in the (z) image. That is, a plurality of images are collected by monitoring (photographing) the possessed object by the camera 25 at a predetermined time, and the motion information of the possessed object at a predetermined time can be collected by performing image recognition processing on those images. After storing the collected motion information of the possessed object in the second area 32b, the process proceeds to step S18.

Next, in step S18, it is determined whether or not the state is abnormal based on the motion information of the possessed object at a predetermined time. Here, the motion information of the possessed object at a predetermined time collected by the image recognition process in step S17 is collated with the motion information of the object requiring attention registered in the first area 32a, and the above is based on the collation result. Is judged.

Here, the description will be made based on the motion information (FIG. 3 (b)) of the object requiring attention registered in the first area 32a. As shown in FIG. 3B, when the object requiring attention is a crowbar, an operation estimated as an abnormal state is registered based on the number of detected crowbars. Specifically, (1) Detect one crowbar as an object to be watched: The history of crowbar movement is based on a fulcrum with a crowbar (one end), and repeats 6 or more reciprocating movements in a movement range of 30 to 150 degrees. In the case of (2) Detecting two crowbars as objects requiring attention: The history of crowbar movement is three or more times in total for two crowbars in a movement range of 30 to 150 degrees based on the fulcrum with the crowbar (one end). In the case of repeated reciprocating operation of, is registered.

When the locking device 21 is locked, the dead bolt 22 is inserted into the receiving hole 23 to form an engaging structure. By destroying the engagement structure between the dead bolt 22 and the receiving portion 23, the suspicious person tries to release the restricted state of the entrance door 13 and make the entrance door 13 openable (that is, pry open). ) In many cases.

Such breaking of the engagement structure between the dead bolt 22 and the receiving portion 23 is often performed by the bar. As an operation at that time, it is assumed that there is a dead bolt 22 in which the tip end portion (one end side) of the bar is inserted into the gap between the entrance 12 and the entrance door 13 and is engaged with the receiving hole 23. The tip portion thereof is brought into contact with the vicinity (the gap between the entrance 12 and the entrance door 13). Then, with the tip portion of the crowbar (the portion abutting the crowbar) as a fulcrum, a force is applied to the tip portion from the other end side of the crowbar using the principle of the lever. At this time, the operation of pushing and pulling the other end side of the crowbar (the operation of applying a force toward the fulcrum) is repeatedly performed with the tip portion as the fulcrum within a predetermined movement range, so that the protruding portion of the entrance door 13 and the dead bolt 22 Force is applied to such things. As a result, the protruding portion of the dead bolt 22 and the like are destroyed, the engagement structure between the dead bolt 22 and the receiving hole 23 is destroyed, and the entrance door 13 can be opened and closed. That is, in the present embodiment, as the operation information of the crowbars (1) and (2), the operation when the entrance door 13 is pried open by the crowbar (the dead bolt 22 is destroyed by the crowbar) is assumed. ..

Usually, when a person other than a suspicious person such as a resident performs work using a crowbar (for example, nailing), the work is often performed with one crowbar. Therefore, if two burs are detected and an operation (reciprocating operation) when the burs pry open is detected, it is very likely that the person is a suspicious person who attempts unauthorized intrusion. Further, since the vandalism with one crowbar is carried out for one locking device 21, for example, when two crowbars are detected, the vandalism can be carried out for the two locking devices 21 at the same time. .. Therefore, in the present embodiment, the motion information of the object requiring attention for presuming the abnormal state is defined for each number of detected crowbars. Specifically, when one crowbar is detected, when two crowbars are detected, the number of reciprocating movements of the crowbar when opening the entrance door 13 is reduced. That is, since it is already a situation to be wary when two burs are detected as possessed objects, in this case, an abnormal state is detected when the total of the reciprocating motion of prying the entrance door 13 with the burs is detected three times. It is configured to be. As a result, the suspicious person can be detected more quickly and the damage can be minimized.

That is, in step S19, it is determined whether or not the operation of the possessed object corresponds to the operation of the object requiring attention registered in the first area 32a as the act of destroying the locking device 21 (dead bolt 22). As a result of collating the motion information of the possessed object with the motion information of the object requiring attention, if the motion information of the possessed object matches the motion information of the object requiring attention, the locking device 21 is destroyed (that is,). , It is an abnormal state), and the process proceeds to step S19. On the other hand, if the motion information of the possessed object does not match the motion information of the registered object requiring attention, it is determined that the locking device 21 has not been destroyed (that is, it is not in an abnormal state), and this process is terminated. do.

In step S19, the security control center 50 is notified. Here, when it is determined that the object is in an abnormal state based on the movement of the object requiring attention, information indicating that the state is abnormal is transmitted to the security control center 50. As a result, the security control center 50 can go to the building 10 which is the transmission destination of the abnormal state, look around the inside and outside of the building 10, catch a suspicious person, and confirm the safety. After notifying the information that it is in an abnormal state, this process ends.

According to the configuration of the present embodiment described in detail above, the following excellent effects can be obtained.

(1) According to the present embodiment, a person is extracted from an image taken by the camera 25, and an object possessed by the person is extracted. Then, when the extracted object corresponds to the object requiring attention, it is determined whether or not it is in an abnormal state based on the operation of the object requiring attention, and a suspicious person is detected.

For example, a suspicious person tries to invade the building 10 by destroying the locking device 21 provided on the entrance door 13. In this case, it is considered that the locking device 21 is destroyed by using some tool (hereinafter referred to as a destruction tool). The destructive tool applies a destructive force to the locking device 21. In other words, the destructive tool is used with a characteristic operation with respect to the locking device 21, so that the locking device 21 is destroyed. According to the present embodiment, paying attention to the characteristic operation of the locking device 21 by such a destructive tool and specifying the operation of the object with respect to the locking device 21 (entrance door 13), the locking device 21 (entrance door 13) is specified. Estimate whether or not the vandalism against 13) has been carried out.

When the resident who has returned home performs a normal unlocking operation, the movement of the object, which is considered to be a destructive act on the locking device 21, is not detected. Further, even a visitor such as a courier may have various situations such as the presence or absence of luggage, but the movement of an object that is considered to be a vandalism to the locking device 21 is not detected. Therefore, according to the present embodiment, a person other than the suspicious person is determined by determining whether or not the person near the entrance door 13 is a suspicious person who intends to invade the building 10 based on the movement of the object which is considered to be a vandalism. It is possible to avoid falsely detecting a person as a suspicious person.

(2) According to the present embodiment, since the suspicious person is detected based on the movement of the object performed at the time of destruction, the abnormal state can be detected before the suspicious person invades the building 10. As a result, it is possible to respond more quickly to the intrusion of a suspicious person, and the damage can be minimized.

(3) According to the present embodiment, a person is recognized in the captured image, and attention is paid to an object possessed by the recognized person. For example, when an object is left near the entrance door 13, the object may move due to wind or the like. In such a case, for example, if the motion of the abandoned object is determined to be a destructive action, it will be erroneously detected as an abnormal state. The act of destroying the locking device 21 (entrance door 13) is performed by a person and an object. According to the present embodiment, since the abnormal state is estimated based on the movement of the object possessed by the person, the vandalism can be detected more reliably and the false detection can be reduced.

(4) According to the present embodiment, an object requiring attention and a specific operation for the object requiring attention are predetermined. As a result, it is possible to more reliably determine whether or not an abnormal state is present based on the extraction of the object requiring attention and the operation of the object requiring attention, and the accuracy of detecting the abnormal state can be improved.

(5) According to the present embodiment, when it is presumed that the server 51 is in an abnormal state, the server 51 is managed to be in the abnormal state by transmitting the warning information indicating the abnormal state to the server 51. The security control center 50 is notified. Therefore, the abnormal state is automatically detected and notified to the security control center 50. By being notified, the security control center 50 can promptly take measures such as going to the building 10 and checking the safety inside and outside the building 10. As a result, it is possible to catch a suspicious person or narrow the time required for the suspicious person to enter the building 10, and the damage can be minimized.

The present invention is not limited to the above embodiment, and may be implemented as follows, for example.

(1) In the above embodiment, the image of a person and an object included in the captured image is recognized by extracting and analyzing the feature amount (color, shape, size, etc.) from the image captured by the camera 25. However, image recognition processing using deep learning may be executed. For example, in deep learning, a multi-layer neural network (DNN) is used. As a multi-layer neural network, for example, there is a CNN (Convolutional Neural Network).

(2) In the above embodiment, the object requiring attention and the motion information of the object requiring attention are predetermined. For example, deep learning is performed using the motion information of the object collected by the image taken by the camera 25. By executing this, a criterion for determining the motion information of the object to be determined to be in an abnormal state may be generated.

(3) In the above embodiment, when the object possessed by the person by image recognition does not correspond to the object requiring attention, the security process is completed without collecting the motion information of the object, but the object possessed by the person. All the operation information of the above may be collected and the data of the operation information may be accumulated.

(4) In the above embodiment, the bar is registered as an object requiring attention that destroys the locking device 21 around the entrance door 13, but the object requiring attention is not limited to this as long as it can destroy the locking device 21. .. For example, in the case of an object that is long and has a flat surface at the tip, the tip portion (one end side) of the flat surface of the object is inserted into the gap between the entrance 12 and the entrance door 13 in the same manner as in the above embodiment. The engagement structure between the dead bolt 22 and the receiving portion 23 can be broken.

(5) In the above embodiment, the security system of the present invention is applied to the entrance 12 provided with the entrance door 13, but for example, another entrance (for example, a self-propelled entrance) or a window provided with an opening / closing body such as a door. The security system of the present invention may be applied to the above. Therefore, in the above embodiment, the operation of destroying the locking device 21 around the entrance door 13 is set as an abnormal state, but for example, the operation of destroying a locking device such as a crescent lock around the window may be set as an abnormal state. good. As a result, even intrusion through a window can be detected, and crime prevention is further enhanced.

In the case of intrusion through a glass window, it is possible to use a bar, but a flat-blade screwdriver is often used. Therefore, in this case, a flat-blade screwdriver is registered as an object requiring attention in the first area 32a. In addition, as motion information (conditions presumed to be an abnormal state) of an object requiring attention, for example, when the tip of a flat-blade screwdriver is directed toward the glass surface of the glass window and the object is moved toward the indoor side of the glass surface. , Etc. are registered.

(6) In the above embodiment, the shooting area SA by the camera 25 is set to be around the entrance door 13, but for example, the shooting area SA up to the entrance porch 15 may be used. In this case, the person who possesses the object corresponding to the object requiring attention can be detected more quickly. For example, the number of repeated operations of the object requiring attention for determining a vandalism (abnormal state) can be reduced. can. As a result, the suspicious person can be quickly detected before the suspicious person invades the building 10, and the damage can be minimized.

(7) In the above embodiment, the controller 31 provided in the building 10 executes the detection (estimation) of the abnormal state based on the image captured by the camera 25. For example, the detection of the abnormal state is security control. It may be executed by the server 51 provided in the center 50. In this case, the building 10 is provided with, for example, a camera 25 as a photographing means, and the image from the camera 25 is transmitted to the server 51 via the communication unit 35. On the server 51 side, based on the image, detection of a person, detection of a possessed object (detection of an object requiring attention), monitoring in the case of an object requiring attention, and estimation of an abnormal state are executed. Then, when an abnormal state is estimated, the server 51 may notify the security control center 50 that the abnormal state is estimated in the building 10 as in the above embodiment, and the same effect as that of the above embodiment can be obtained. can get.

(8) In the above embodiment, the security system for the residential building 10 as the structure has been described, but the structure to which the security system of the present invention is applied is not limited to the residential building 10. For example, it may be a structure such as a vehicle, a commercial building, a warehouse, or a safe.

10 ... Building, 12 ... Entrance as an opening, 13 ... Entrance door as an opening and closing body, 21 ... Locking device, 25 ... Camera as a photographing means, 31 ... Extracting means, operation information collecting means, state estimating means Controller, 31a ... 1st area, 32 ... Object database requiring attention, 50 ... Security management center, 51 ... Server, SA ... Shooting area.

Claims (4)

A security system used for crime prevention of a structure including an opening / closing body for opening / closing an opening and a locking device for locking the opening / closing body.
A photographing means for photographing a predetermined range in the vicinity of the opening / closing body as a photographing area, and
An extraction means for extracting an object from an image captured by the photographing means,
An motion information collecting means that tracks the motion of an object extracted by the extraction means and collects motion information of the object, and an motion information collecting means.
When the object extracted by the extraction means is a long object having a flat surface at the tip, the object identification means for identifying the object as an object requiring attention and the object identification means.
A state estimating means for presuming an abnormal state when the motion information collecting means collects motions of moving back and forth within a predetermined angle range a predetermined number of times with the tip of the object requiring attention as a fulcrum. When,
A security system characterized by being equipped with. The crime prevention system according to claim 1, wherein the extraction means extracts an image of a person from an image taken by the photographing means, and then extracts an object possessed by the person. The state estimation means is provided with a determination standard generating means for generating a determination criterion of an abnormal state by executing deep learning using the information collected by the operation information collecting means. Or the security system according to 2 . The security system according to any one of claims 1 to 3 is provided.
The photographing means is provided corresponding to each of the plurality of opening / closing bodies.
It is equipped with a management center that manages the security system.
When the state estimation means presumes that the predetermined opening / closing body is in an abnormal state, the security system notifies the management center of the warning information indicating the abnormal state by associating it with the predetermined opening / closing body. A crime prevention management system that is characterized by doing so.

Images