KR101470316B1 - Closed-circuit television system of sensing risk according to the characteristic of sensing area and method thereof - Google Patents

Abstract

Disclosed are a closed-circuit television (CCTV) system for sensing a risk according to characteristics of a sensed area and a method thereof. The CCTV system of the present invention comprises: a CCTV camera configured to capture images of a specific photographing area; a CCTV image storage module configured to store the images captured by the CCTV camera; an event area setting database configured to register and store each area photographed by three-axial coordinates of pitch, yaw, and roll of the CCTV camera after classifying and setting the area into multiple event areas; an object recognition module configured to automatically recognize the type of object included in a received image based on the size and motion of the object after receiving the captured image from the CCTV camera, or automatically recognize the type of object included in an image stored in the CCTV image storage module based on the size and motion of the object; an object risk determination module configured to query an event area where the object is recognized by referring to the event area setting database and determine a risk based on characteristics of the queried event area and the type and the motion of the object recognized in the event area; and a siren output module configured to generate and output a warning siren or an emergency siren according to the risk of the object in the event area determined by the object risk determination module.

Description

Technical Field [0001] The present invention relates to a risk-sensitive CCTV system and a method thereof, and more particularly to a risk-

The present invention relates to a CCTV system, and more particularly, to a risk-sensitive CCTV system and its method according to characteristics setting of a shooting region, and more particularly, to a CCTV system and method for dividing a shooting region into a plurality of detailed event regions, The present invention relates to a CCTV system and a method thereof, in which an object in a corresponding event area is differently interpreted according to a characteristic of an event area, and the danger according to the object is quickly and accurately detected in advance.

Closed-circuit television (CCTV) systems are increasingly being installed for theft prevention and intrusion prevention. In the parks and alleys, the complaints about the installation of CCTV are flooding, and the CCTV system plays a big role in detecting crime prevention and natural disasters.

The biggest disadvantage of this existing CCTV system is that if crime, theft or intrusion occurs, it is mainly a post - In other words, it is not enough to cope with theft / intrusion by the CCTV system immediately.

Of course, a 24-hour monitor can monitor major security facilities, but it is almost impossible to monitor many CCTV cameras. CCTVs installed in houses and apartment complexes are mostly only after-effects, and there are no monitors installed.

As such, CCTV cameras installed in apartment complexes, houses, and other large security facilities are required to detect movements in real time through the images of CCTV cameras.

On the other hand, a conventional object recognition technique in an image is disclosed in the following documents.

Registered Patent Publication No. 10-1176743 discloses a technique for recognizing an object in a CCTV moving image. Japanese Patent Application Laid-Open No. 10-1176743 discloses a technique of previously generating a still image for a specific object and confirming that the CCTV moving image is the same object in contrast to an object in the captured image.

Japanese Unexamined Patent Application Publication No. 2002-0021459 discloses a technology for recognizing and recognizing detailed movements such as dribbling, running, walking, standing, and kick by understanding the movement of each player in a soccer game video.

As described above, the conventional object recognition technology can recognize not only the object but also the detailed movement of the object. However, in the conventional CCTV system, a system for predicting or preventing a crime through the recognition or movement of such an object is not yet known nor shared.

It is needless to say that the algorithm of the above-mentioned document can not be applied as it is. That is, the conventional CCTV system can also be prepared by using a captured image of an object, that is, a person, as in the configuration disclosed in Patent Registration No. 10-1176743, or whether the person is a crime to commit, I can not grasp it at all. In other words, it is impossible to anticipate or prevent crime or invasion in advance of the behavior of each object in real time.

Therefore, in order to solve the biggest disadvantage of the CCTV system at present, there is an urgent and urgent need for effective measures to prevent and prevent crime in real time.

10-1176743 (2002-0021459)

It is an object of the present invention to provide a risk-sensitive CCTV system according to characteristics setting of a shooting area.

It is another object of the present invention to provide a method for detecting a risk of a CCTV system according to characteristics setting of a shooting area.

According to an aspect of the present invention, there is provided a risk-monitoring CCTV system comprising: a CCTV camera for capturing an image of a predetermined photographing area; A CCTV image storage module for storing images photographed by the CCTV camera; An event area setting database for separately registering and storing each photographing area for each of three-axis coordinates of the pitch, yaw, and roll of the CCTV camera into a plurality of event areas; Receiving the image photographed from the CCTV camera, automatically recognizing the object included in the received image by the size and motion of the object, or storing the object included in the image stored in the CCTV image storage module into the object size And an object recognition module for automatically recognizing the type of the object by motion; An object risk determination module for an event area for inquiring the event area in which the object is recognized by referring to the event area setting database and determining a risk according to the characteristics of the detected event area and the type and movement of the object recognized in the event area, ; And a siren output module for generating and outputting an alarm siren or an emergency siren in accordance with the risk of the object determined by the object risk determination module for each event area.

At this time, the object may be any one of a person, a vehicle, and an animal.

The event area is an area divided into a plurality of polygonal areas in one rectangular photographing area, and is preset in any one of an intrusive area and an intruding area, and at least one of an entrance area, a passage area, It is preferable that the characteristic region is set in advance.

The event risk management module for each event area may be configured such that an event area in which the object is recognized is set as an intrusive area and if the motion of the object does not correspond to a predetermined motion of the property area, It can be configured to judge it as a risk.

If the object is a car, the object danger zone determination module may move the object at a predetermined speed or more in a parking area, or if the object is a person, when the object is roaming, stopping, or invading in an entrance area, . ≪ / RTI >

According to another aspect of the present invention, there is provided a method of detecting a risk of a CCTV system, the method comprising: capturing an image of a predetermined region by a CCTV camera; The CCTV image storage module storing an image captured by the CCTV camera; The object recognition module receives the image photographed from the CCTV camera and automatically recognizes the object included in the received image by the size and motion of the object Automatically recognizing the object included in the image stored in the CCTV image storage module by the size and motion of the object; An object danger zone determination module for each event area refers to an event area in which the object is recognized by referring to the event area setting database; Determining a risk according to the characteristics of the searched event area and the type and movement of the object recognized in the event area; And generating and outputting an alarm siren or an emergency siren according to the risk of the object determined by the object-area-object-risk-determination module in the siren output module.

At this time, the object may be any one of a person, a vehicle, and an animal.

The event area is an area divided into a plurality of polygonal areas in one rectangular photographing area and is preset in any one of an invasive area and a non-invasive area, and at least one of an entrance area, a passage area, and a parking area As shown in FIG.

The step of determining the risk according to the characteristics of the inquired event area and the type and movement of the object recognized in the event area may include determining whether the event area in which the object is recognized is an intrusiveness area And may be configured to determine the object as dangerous if the motion of the object does not match a predetermined motion of the property area.

The step of determining the risk according to the characteristics of the searched event area and the type and movement of the object recognized in the event area may include determining whether the object is a car, Or if the object is a person, it may be configured to determine that the object is dangerous when it is roaming, stopping, or invading in an entry area that is an intrusive area.

According to the risk detection CCTV system and the method according to the characteristic setting of the shooting area, the CCTV image is divided into / divided into event areas of polygons, and each event area is set as an intrusive area or an intrusive area, A fence area, and a passage area. Thus, it is possible to more effectively predict an object in the CCTV image and the risk associated with the movement, and to prevent intrusion.

Also, the object in the CCTV image is configured to recognize the object by its size, motion pattern, motion speed, feature point extraction, and register the suspicious motion corresponding to each event area in advance to monitor the motion, It is possible to more precisely and precisely recognize the danger caused by the risk.

In CCTV images, there may be various patterns such as movement of leaf or tree branches or movement of bees as well as dynamic objects. Therefore, by filtering these various motions using an event area and filtering them, The risk detection level by the event area algorithm is very high.

1 is a block diagram of a risk-sensitive CCTV system according to an exemplary embodiment of the present invention.
2 is a flowchart illustrating a method of detecting a risk of a CCTV system according to an exemplary embodiment of the present invention.

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 to the concrete inventive concept.

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.

Like reference numerals are used for like elements in describing each drawing.

The terms first, second, A, B, etc. may be used to describe various elements, but the elements should not be limited by the terms. The terms are used only for the purpose of distinguishing 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 when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, .

On the other hand, when an element is referred to as being "directly connected" or "directly connected" to another element, it should be understood that there are no other elements in between.

The terminology used in this application is used only to describe a specific embodiment and is not intended to limit the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise.

In the present application, the terms "comprises" or "having" and the like are used to specify that there is a feature, a number, a step, an operation, an element, a component or a combination thereof described in the specification, But do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or combinations thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

Terms such as those defined in commonly used dictionaries are to be interpreted as having a meaning consistent with the contextual meaning of the related art and are to be interpreted as either ideal or overly formal in the sense of the present application Do not.

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram of a risk-sensitive CCTV system according to an exemplary embodiment of the present invention.

Referring to FIG. 1, a CCTV system 100 according to an exemplary embodiment of the present invention includes a closed circuit television camera 110, A CCTV image storage module 120, an event area setting database 130, an object recognition module 140, an object risk determination module 150, and a siren output module 160 Lt; / RTI >

The CCTV system 100 is configured to recognize a dynamic object such as a person, a vehicle, a dog, a cat, and the like in real time in the CCTV image and grasp the movement thereof. Recognition of these objects can be done in various ways by object size, movement, and minutiae extraction of object. The weights of the recognition or the performance of the algorithm for recognition can be set variously.

The CCTV system 100 recognizes the type of the object itself through the size and motion of the object, and for the recognized object, predicts an abnormal symptom of the object or possibility of invasion through the movement again.

In order to further increase the accuracy of the prediction algorithm, the CCTV system 100 divides / divides a CCTV image into a plurality of event regions and assigns characteristics to the respective event regions. For example, each event area is set as an intrusive area and an intrusive area. The invasive area can be a security zone such as entrance, fence, and large safe in CCTV video. The intruder zone can not be invaded, but it is an area where intrusion is unnecessary. CCTV It means the part captured by the image.

Movement of non-intrusive areas such as rivers, plazas, trees, etc., can be judged very low even if object recognition is incomplete. That is, there may be various patterns such as movement of leaves or branches or movement of bees. Therefore, if the various motions are filtered and filtered using the event area, the risk detection level can be very high and the calculation amount can be reduced, .

Such an event area may be set differently for each of the images each time the direction of the CCTV camera 110 is changed. That is, the event area can be divided / partitioned in the first direction video, the second direction video, and the third video direction, respectively.

Hereinafter, the detailed configuration will be described.

The CCTV camera 110 may be configured to photograph an image of a predetermined photographing area. At this time, a plurality of photographing areas can be preset by changing the direction of the CCTV camera 110 in the photographing area. The CCTV camera 110 can be configured to be able to drive in three axial directions of pitch, yaw, and roll.

The CCTV image storage module 120 may be configured to store images photographed by the CCTV camera module 110.

The event area setting database 130 divides and sets each photographing area for each of the three axis coordinates of the pitch, yaw and roll of the CCTV camera 110 into a plurality of event areas Registration and storage. That is, the shooting area corresponding to each shooting direction is set in advance as a plurality of event areas.

Here, the event area is an area divided into a plurality of polygonal areas in one rectangular photographing area, and can be set in advance as any one of an intrusive area and an intolerable area. Each event area may be set as a characteristic area including at least one of an entrance area, a passage area, and a parking area.

That is, the event area is set to one of the intrusive area and the non-intrusive area, and is also set to any one of the characteristic areas.

The object recognition module 140 receives an image photographed from the CCTV camera 110 and automatically recognizes the object included in the received image by the size and motion of the object, The object included in the image stored in the CCTV image storage module 120 can be automatically recognized by the size and motion of the object.

Here, the object may be composed of any one of a person, a vehicle, and an animal.

Objects frequently appearing in the CCTV image are preferably registered in the object recognition module 140 in advance. This is because it is not necessary for a fish or bird not to be admitted to be registered.

Also, it is preferable that the object recognition module 140 registers the object size, motion pattern, motion speed, and the like in advance based on the past haunting image.

The object recognition module 140 can primarily grasp the type of the object, such as a person or a person, through the size of the object displayed on the image, the speed or the pattern of the movement. In other words, since a person, a vehicle, and a dog each have different sizes appearing in an image, the object can be grasped primarily according to its size, and the movement speed of the object is different from that of a person, The object can be recognized or grasped.

Secondarily, the object recognition module 140 can grasp the feature points extracted from the shape of the object to grasp the type of the object. At this time, feature point extraction algorithms such as Scale Invariant Feature Transform (SIFT) algorithm and SURF (Speed Up Robust Feature) algorithm can be used.

It is desirable that both the first and second algorithms are simultaneously performed. Although the accuracy of the first algorithm is lower than that of the second algorithm, an improved algorithm such as comparing the size of objects registered in the past image The object can be estimated faster.

Meanwhile, when the object is recognized, the object recognition module 140 may be configured to determine whether the movement of the object is past, whether the object is moving, stopping, or intruding.

The object recognition module 140 judges that a person passes by, a person stops, a person walks, or the like by combining an object and its motion in a pair.

The object-risk-per-event-risk determination module 150 is configured to determine the risk of an object by associating an object determined by the object recognition module 140 and a motion pair with an event area in which the object exists.

That is, the object-risk-per-event-risk determination module 140 is configured to determine a risk according to the characteristics of the inquired event area and the type and movement of the object recognized in the corresponding event area.

Here, the object-risk-per-event-risk determination module 150 may be configured to query an event area of an object with reference to the event area setting database 130. [

If the event region in which the object is recognized is set in advance as an intrusive region and the motion of the object does not match a predetermined motion of the characteristic region, the object risk determination module 150 for each event region determines that the object is dangerous .

For example, it can be judged to be very dangerous when a combination such as a long-term texting or a long-term stopping of a person is recognized in an intruding area such as an entrance area or a fence area. This is because a combination of an object and a motion that does not fit in the characteristic region is displayed in the exit area or the fence area for a long time or for a long time.

However, even if the object and its movement are suspicious combinations such as a person hanging for a long time or a person hanging for a long time, the event area in which the person exists is not limited to a security facility, an entrance area, a square unrelated to a fence area, It can be judged that there is no risk.

The siren output module 160 may be configured to generate and output an alarm siren or an emergency siren according to the risk of the object determined by the object area risk object determination module 150. [

2 is a flowchart illustrating a method of detecting a risk of a CCTV system according to an exemplary embodiment of the present invention.

Referring to FIG. 2, a closed circuit television (CCTV) camera 110 captures an image of a predetermined photographing area (S101).

Next, the CCTV image storage module 120 stores the image photographed by the CCTV camera 110 (S102).

Next, the object recognition module 130 receives the image photographed from the CCTV camera 110 and automatically recognizes the object included in the received image by the size and motion of the object The object included in the image stored in the CCTV image storage module 120 is automatically recognized by the size and motion of the object (S103).

Here, the object may be any one of a person, a vehicle, and an animal.

Next, the object-risk-per-event-risk determination module 140 refers to the event area setting database 140 to inquire the event area in which the object is recognized (S104).

Here, the event area means an area divided into a plurality of polygonal areas in one rectangular shooting area.

The event area may be preset to one of the intruder area and the intruder area.

The event area may also be set as a characteristic area such as an entrance area, a passage area, and a parking area. The characteristic area may be additionally set in addition to the entrance area, the passage area, and the parking area, depending on the characteristics of the facility.

Next, the object-risk-per-event-risk determination module 150 determines the risk according to the characteristics of the event area and the type and movement of the object recognized in the event area (S105).

Here, if the event area in which the object is recognized is preset as an intruding area and the movement of the object does not match the predetermined movement of the property area, the object area risk event determination module 150 determines that the object is dangerous .

In addition, when the object is a vehicle, the object risk determination module 150 for each event area may determine whether the object is moving at a speed higher than a predetermined speed in the parking area or if the object is a person, It can be configured to judge it as dangerous.

Next, the siren output module 160 generates and outputs an alarm siren or an emergency siren according to the risk of the object determined by the object-area-based object risk determination module 150 (S106).

It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or scope of the invention as defined in the following claims. There will be.

110: CCTV camera 120: CCTV image storage module
130: Event area setting database 140: Object recognition module
150: object danger judgment module 160: siren output module

Claims (4)

A CCTV camera (closed circuit television camera) for capturing an image of a predetermined photographing area;
A CCTV image storage module for storing images photographed by the CCTV camera;
An event area setting database for separately registering and storing each photographing area for each of three-axis coordinates of the pitch, yaw, and roll of the CCTV camera into a plurality of event areas;
Receiving the image photographed from the CCTV camera, automatically recognizing the object included in the received image by the size and motion of the object, or storing the object included in the image stored in the CCTV image storage module into the object size And an object recognition module for automatically recognizing the type of the object by motion;
An object danger judgment module for an event area for inquiring an event area in which the object is recognized by referring to the event area setting database, and determining a risk according to the characteristics of the searched event area and the kind and movement of the object recognized in the event area;
And a siren output module for generating and outputting an alarm siren or an emergency siren in accordance with the risk of the object determined by the object risk determination module for each event area,
The object comprising:
A person, a vehicle, or an animal,
Wherein the event area comprises:
Wherein the plurality of polygon areas are divided into a plurality of polygon areas in one rectangular photographing area and are preset in any one of an invasive area and an inviolable area and are stored in advance as characteristic areas including at least one of an entrance area, Is set,
Wherein the object-risk-per-
Wherein if an event area in which an object is recognized is preset as an intrusive area and motion of the object does not match a predetermined motion of the property area,
Wherein the object-risk-per-
Stop or infiltrate in an entry area, which is an intruding area, when the object is a person or moves to a predetermined speed or more in a parking area when the object is a vehicle,
In the object recognition module, objects frequently appearing in a CCTV image are registered in advance, and the size, motion pattern, and motion speed of the object are registered in advance based on a past haunting image,
Wherein the object recognition module extracts feature points using a Scale Invariant Feature Transform (SIFT) algorithm and a SURF (Speed Up Robust Feature) algorithm in the shape of an object, Hazard detection CCTV system.
delete delete delete

Images