KR101858396B1 - Intelligent intrusion detection system - Google Patents

Abstract

The present invention relates to a functional intrusion detection system, and more particularly, to a camera module for capturing an image of a surveillance area including a security facility and monitoring an object entering the surveillance area; And a control unit for setting a virtual fence on the image of the photographed surveillance area, identifying an object as an intruder, generating an alarm according to the identification result, and transmitting a control signal to the camera module to track the movement path and the motion of the object Module. Therefore, the system grasps the identity or movement of the intruding object in the surveillance area, and provides an alarm generating effect according to the situation.

Description

[0001] INTELLIGENT INTRUSION DETECTION SYSTEM [0002]

The disclosed technique relates to an intelligent intrusion detection system that uses a virtual fence to prevent intrusion into a security facility.

In buildings such as schools and government offices that protect inside the building or distinguish them from outside people, physical structures such as walls and barricades are used to clarify the boundary between inside and outside of the building. In recent years, however, the efficiency of these physical structures is not so high, and it has been changed to be equipped with other kinds of alarm systems such as a lowering of the height or a cc TV due to a bad effect on aesthetics. In accordance with this change, the physical structure has not changed at all, and ccTV and crime prevention systems are being deployed everywhere.

Meanwhile, such a crime prevention system generally detects an intruder by using a sensor that can cover a predetermined area around a building or a security facility. For example, when an infrared sensor and a cc TV are placed around a building to detect a human body, and the sensor detects heat corresponding to the human body, the image is acquired by controlling the cc tv in that direction. Can be understood. Of course, it is also possible to construct a system that detects fingerprints and faces for each compartment inside the building.

However, this approach has the disadvantage that the administrator must always wait near the site. For example, if an alarm is received by the control panel through the control panel, the user can manually control the CCTV or view the image of the CCTV displayed on the screen of the control panel to check whether the user is an intruder or an internal person. Although this method can provide high accuracy for intrusion detection, it shows a somewhat low efficiency in determining whether the object photographed in the image is an intruder parameter.

Korean Patent Laid-Open No. 10-2013-0130958 entitled " Electronic fence system " is known as a conventional technology for detecting an intrusion using a sensor instead of a physical structure. Referring to the related art, there is disclosed a technique for detecting an intruder entering a surveillance area using a virtual fence. However, there is a problem that the administrator has to directly filter the distinction between the intruder and the passing person.

The disclosed technique is to provide an intelligent intrusion detection system that uses a video-based virtual fence to prevent intrusion into a security facility and to distinguish an intruder from the system.

According to a first aspect of the present invention, there is provided a camera module for capturing an image of a surveillance area including a security facility and monitoring an object entering the surveillance area, And a control module for identifying the intruder as an object, generating an alarm according to the identification result, and transmitting a control signal to the camera module to track the movement path and the motion of the object, in an intelligent intrusion detection system .

Embodiments of the disclosed technique may have effects that include the following advantages. It should be understood, however, that the scope of the disclosed technology is not to be construed as limited thereby, since the embodiments of the disclosed technology are not meant to include all such embodiments.

According to an embodiment of the disclosed technology, the intelligent intrusion detection system has an advantage that the system directly detects an intruding object in the surveillance area and generates an alarm.

Also, there is an effect of accurately identifying the identity or type of an object by using a plurality of images photographed by different kinds of cameras.

In addition, there is an advantage of setting the advance alarm area and tracking the movement and movement direction of the object in the area to generate an alarm accurately.

1 is a block diagram of an intelligent intrusion detection system in accordance with one embodiment of the disclosed technology.
2 is a flowchart illustrating an operation of an intelligent intrusion detection system according to an embodiment of the disclosed technology.
3 is a diagram illustrating the creation of a virtual fence in accordance with one embodiment of the disclosed technique.
4 is a view illustrating setting of a surveillance area around a security facility according to an embodiment of the disclosed technology.
5 is a diagram illustrating a prior alert area in accordance with one embodiment of the disclosed technique.

While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail. It should be understood, however, that the invention is not intended to be limited to the particular embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.

The terms first, second, A, B, etc., may be used to describe various components, but the components are not limited by the terms, but may be used to distinguish one component from another . For example, without departing from the scope of the present invention, the first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component. And / or < / RTI > includes any combination of a plurality of related listed items or any of a plurality of related listed items.

It is to be understood that the singular < RTI ID = 0.0 > terms < / RTI > used herein should be interpreted to include a plurality of representations unless the context clearly dictates otherwise. And "comprises ", when used in this specification, specify the presence of stated features, numbers, steps, operations, elements, parts, or combinations thereof, Or combinations thereof, as a matter of course.

Before describing the drawings in detail, it is to be clarified that the division of constituent parts in this specification is merely a division by main functions of each constituent part. That is, two or more constituent parts to be described below may be combined into one constituent part, or one constituent part may be divided into two or more functions according to functions that are more subdivided.

In addition, each of the constituent units described below may additionally perform some or all of the functions of other constituent units in addition to the main functions of the constituent units themselves, and that some of the main functions, And may be carried out in a dedicated manner. Accordingly, the presence or absence of each component described in this specification should be interpreted as a function.

1 is a block diagram of an intelligent intrusion detection system in accordance with one embodiment of the disclosed technology. And FIG. 2 is a flowchart illustrating an operation of the intelligent intrusion detection system according to an embodiment of the disclosed technology. Referring to FIGS. 1 and 2, the intelligent intrusion detection system 100 includes a camera module 110 and a control module 120. The control module 120 transmits an alarm to the administrator terminal 130.

The camera module 110 captures an image of the surveillance area. In one embodiment, the camera module 110 may continuously photograph an image including a surveillance area set around the security facility according to a predetermined period. Here, the surveillance area is formed so as to surround the perimeter of the security facility as in a normal security system. For example, if the security facility 400 exists in the center as shown in FIG. 4, the surveillance area 401 may surround the certain radius of the surveillance area 401. The area adjacent to the security facility may be set as the surveillance area 401 and the area outside the security area may be set as the surveillance area 402 based on the specific boundary line. Of course, the surveillance area 401 and the non-surveillance area 402 are not physically displayed on the image but are preferably used as a kind of guide line for setting a virtual fence or a prior alert area.

Meanwhile, the camera module 110 includes different kinds of cameras. For example, a plurality of cameras may be provided, some of which may be a thermal camera 111 and others may be an optical camera 112. The camera module 110 may convert the captured images into data that can be used to determine what the object is.

In one embodiment, the camera module 110 may match two images captured from different types of cameras using a predetermined algorithm. If the heterogeneous camera is assumed to be the thermal camera 111 and the optical camera 112 as described above, the thermal imaging camera 111 can take an image of the temperature distribution of the object, It is possible to take an image of the external shape of the camera. It is difficult to specify the object by only the image including the temperature taken by the thermal imaging camera 111. The image taken by the optical camera 112 may also be photographed differently depending on the external environment Since there is a high possibility that noise or distortion occurs in an image due to various factors due to characteristics of an image photographed by the optical camera 112, it is difficult to specify an object with only one image. However, if the information on the temperature distribution and appearance of an object are merged into one, it is possible to improve the accuracy of the object by grasping the object through a single image.

In addition, in another embodiment, the camera module 110 combines the boundary line estimated as an object in the image photographed by the thermal imaging camera 111 and the boundary line estimated as an object in the image photographed by the optical camera 112, Can be narrowed to a specific prime number from an unspecified number. In addition, it is possible to convert the data into data that can be used to determine what the photographed object is by matching information on various types of objects stored in advance.

The control module 120 receives the matching image of the camera module 110 and identifies whether the object is an intruder or not. In one embodiment, the control module 120 continuously receives an image of an object from the camera module 110, performs a deep learning operation on the continuous image to extract the feature points, and stores the feature points and the pre- The object information may be compared to determine whether the object is a chip particle or a wild animal. In one embodiment, the control module 120 includes a feature point extractor 122 and a database 123. The database 123 stores a plurality of information about a specific object in advance, The minutiae point extraction unit 122 extracts the minutiae points of the object included in the image and matches the plurality of object information stored in the database 123 to find the one having a high degree of similarity.

In another embodiment, the control module 120 receives a continuous image of the object from the camera module 110 for a predetermined time, performs a deep learning operation on the continuous image to extract the feature point, The object information stored in the database 123 may be compared to determine whether the object is a chip particle or a wild animal. The control module 120 can extract the feature points of the object, extract characteristic portions of the size and the appearance of the object, and compare the extracted characteristic points with the pre-stored object information, thereby identifying what the object is photographed.

Here, the control module 120 may determine that the object is an intruder if the identified object is a person. In general, wildlife and non-living things often come from the outside into the surveillance area simply by environmental factors, rather than causing serious harm to security facilities. However, since it is not preferable from the viewpoint of efficiency, the disclosed technology operates so that the administrator can check if the object is recognized as an intruder.

Meanwhile, it is preferable that the control module 120 receives the image of the surveillance area before the camera module 110 monitors the object. It is also possible to set a virtual fence by receiving a plurality of coordinates from the manager on the previously received image. 3, the control module 120 receives information on a plurality of coordinates. In one embodiment, when an administrator inputs a specific coordinate through the interface 121, the user can receive this value and use it as information for generating a virtual fence. Here, the plurality of coordinates can be input at the same position as the position at which the alarm sensor is disposed.

The control module 120 receives the image of the specific region previously photographed by the camera module before the camera module 110 photographs the object, arranges a virtual boundary line on the image of the specific region, Can be generated. The virtual fence is generated based on the coordinates input by the administrator of the system 100 through the interface 121. At this time, an image for a specific area including the security facility 400 is displayed on the screen, Can be input.

Of course, any origin can be input as coordinates regardless of the position of the sensor. Then, the control module 120 generates the virtual fence 301 by interpolating between the coordinates. For example, as in the spline interpolation method, a virtual fence 301 can be generated by connecting a coordinate point and another coordinate point with a smooth curve. This is because when a coordinate point is simply connected by a straight line, a large number of coordinate points are required when an administrator of the system 100 inputs these coordinates through the interface 121, This is because the virtual fence 301 can be generated by inputting the coordinate point of the number.

When the virtual fence 301 is generated as described above, the system 100 sets a region adjacent to the security facility 400 as a surveillance region 401 based on the virtual fence 301 as shown in FIG. 4 , And sets the outer area to the non-monitoring area (402). When the sensor is out of the range where the object can be detected, the monitoring area 401 may generate a shadow interval according to the object detection, so that it is preferable to generate the virtual fence 301 at a proper distance from the security facility 400.

On the other hand, when the control module 120 identifies an intruder as an intruder, the control module 120 transmits a control signal to the camera module 110 to track the object in order to check the movement or direction of the object. In one embodiment, when the control module 120 distinguishes the surveillance area from the non-surveillance area using the virtual fence, the camera module 110 can access the surveillance area from the non- It is possible to track the movement path of the object.

On the other hand, the control module 120 can set a part of the area adjacent to the virtual fence in the surveillance area as the advance alarm area. For example, as shown in FIG. 5, the pre-monitoring area 1 501 and the pre-monitoring area 2 502 can be further set based on the virtual fence. When either or both of the two are inputted into the system 100 as the advance warning area and an object is entered in this area, the camera module 110 can be controlled to track the movement path and movement of the object.

In one embodiment, the camera module 110 captures an image of the surveillance area according to step S210, and transmits the captured image according to step S220. When the control module 120 receives the image, the control module 120 identifies the object in the received image according to step S230. If the identified object is an intruder, an alarm may be generated in step S240. In step S250, a control signal may be transmitted to the camera module 120 in step S250 in order to track the movement or direction of the object identified as the intruder.

If the object has not been identified as an intruder, the control module 120 may generate an alert if the security facility moves in the direction of the location of the security facility in the advance alarm zone and enters the monitoring zone.

On the other hand, in generating an alarm, the control module 120 generates an alarm to the intruder and the terminal 130 of the administrator to access the security facility. In one embodiment, an audible alarm may be output to a sensor or speaker disposed on site to allow an intruder to access the security facility. The administrator terminal 130 may transmit an alarm message indicating that an intruder has been detected. Accordingly, the system 100 itself has an advantage of increasing the accuracy of the security system and increasing the efficiency of the security manager by identifying whether the object is an intruder and generating an alarm according to the situation.

Although the intelligent intrusion detection system according to an embodiment of the disclosed technology has been described with reference to the embodiments shown in the drawings in order to facilitate understanding, it is to be understood that those skilled in the art will appreciate that various modifications, It will be appreciated that other equivalent embodiments are possible. Accordingly, the true scope of protection of the disclosed technology should be determined by the appended claims.

100: intelligent intrusion detection system 110: camera module
111: thermal imaging camera 112: optical camera
120: control module 121: interface
122: feature point extracting unit 123:
130: administrator terminal

Claims (9)

A camera module for capturing an image of a surveillance area including a security facility and monitoring an object entering the surveillance area; And
A control module for setting a virtual fence on an image of the photographed surveillance area, generating an alarm according to an identification result of identifying the object as an intruder, and transmitting a control signal to the camera module to track a movement path and a movement of the object, ≪ / RTI >
Wherein the camera module captures an image of the surveillance region before capturing the object and transmits the image to the control module,
Wherein the control module generates a virtual boundary line on the image of the surveillance area by interpolating the plurality of coordinates with a curve when the plurality of coordinates are input on the surveillance area using an interface provided on a predetermined screen, A fence is set, a part of the area adjacent to the virtual fence is classified as a pre-alarm area, and when the object intrudes into the pre-alarm area, the camera module is controlled to track the movement path and movement of the object, Generates the alert when the security device moves in a direction in which the security facility is located in the advance alarm area. The method according to claim 1,
The camera module includes two heterogeneous cameras, and captures the object by controlling the heterogeneous camera. 3. The method of claim 2,
Wherein the camera module matches the two images photographed from the heterogeneous camera using a predetermined algorithm and acquires an image of the object in the matched image. The method according to claim 1,
Wherein the camera module comprises a thermal imaging camera and an optical camera. delete delete delete delete The method according to claim 1,
Wherein the control module continuously receives an image of the object from the camera module and performs a deep learning operation on the continuous image to extract feature points and compares the feature points with pre-stored object information in a database, An intelligent intrusion detection system that determines whether a chip is a particle or a wild animal.

Images