KR101524922B1 - Apparatus, method, and recording medium for emergency alert - Google Patents

Abstract

The present invention relates to an apparatus, method and recording medium for emergency notification. An emergency situation notification device connected to a surveillance camera according to the present invention includes a database unit in which detection target information and behavior pattern information are matched; A determination unit for analyzing an image of a subject photographed through the surveillance camera and determining a shape and a behavior pattern of each subject; A selection unit configured to compare a shape of each of the objects determined by the determination unit with the detection subject information to select a subject to be sensed; And a warning alarm unit for generating a warning alarm signal when the behavior pattern of the sensing object selected by the selection unit matches the behavior pattern information matched in the database.

Description

[0001] APPARATUS, METHOD, AND RECORDING MEDIUM FOR EMERGENCY ALERT [0002]

The present invention relates to an apparatus, a method, and a recording medium for notifying an emergency situation, and more particularly, to an apparatus, a method, and a recording medium for detecting and warning an emergency situation according to a specific action pattern of a particular person.

CCTV (Closed Circuit Television) is being installed to prevent damage from theft and robbery, as well as private homes.

However, in general, if CCTV is installed, CCTV images are checked 24 hours a day, so that outsiders can steal money from the counter or unauthorized intruders can monitor the situation to prevent damage.

In this way, if a person does not monitor the CCTV video 24 times and only records it, it may be possible to confirm the crime situation after the fact, but there is no effect preventing the accomplishment of the purpose of the crime.

However, it is very costly for each individual to hire a separate monitoring staff, or to join an expense specialist, where the monitoring staff resides.

It is not cost effective to monitor the CCTV installed in the shops of countless members directly by the monitoring personnel in the viewpoint of the expense professional company where the monitoring staff resides.

Published Patent No. 10-2006-0074316

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and it is an object of the present invention to provide a security system that can automatically detect the occurrence of an emergency situation and generate a warning alarm signal And to provide a recording medium.

According to an aspect of the present invention, there is provided an emergency notification apparatus connected to a surveillance camera, the emergency notification system comprising: a database unit in which detection target information and behavior pattern information are matched; A determination unit for analyzing an image of a subject photographed through the surveillance camera and determining a shape and a behavior pattern of each subject; A selection unit configured to compare a shape of each of the objects determined by the determination unit with the detection subject information to select a subject to be sensed; And a warning alarm unit for generating a warning alarm signal when the behavior pattern of the sensing object selected by the selection unit matches the behavior pattern information matched in the database.

In order to accomplish the above object, the present invention provides a method of controlling an emergency situation notification device having a database matching detection target information and behavior pattern information and connected to a surveillance camera, the method comprising: And determining a shape and a behavior pattern of each of the objects; Comparing the shape of each of the determined objects with the detected object information to select an object to be sensed among the objects; And generating a warning alarm signal when the selected behavior pattern of the sensing object matches the behavior pattern information matched in the database.

As described above, according to the present invention, the emergency situation notification device connected to the surveillance camera automatically analyzes the shot image taken by the surveillance camera and automatically determines whether an emergency situation has occurred.

Particularly, in judging whether or not an emergency situation has occurred, by taking into consideration a specific behavior pattern previously registered for a specific registered person, an unregistered person performs an action corresponding to the registered behavior pattern, or a previously registered specific person is registered It is possible to filter out a case in which the operation is not performed, and the error rate of the emergency occurrence occurrence alarm can be greatly reduced.

Even in the case of performing an operation corresponding to a specific registered behavior pattern of a particular registered person, by generating an alarm according to the occurrence of an emergency situation in consideration of a time zone, it is possible to generate a warning alarm according to a specific behavior pattern different between day and night, It is possible to increase the accuracy of a situation occurrence warning alarm.

1 is a schematic block diagram of an entire security system including an emergency situation notification device according to an embodiment of the present invention,
Fig. 2 is a functional block diagram of the emergency situation notification device of Fig. 1,
3 is a control flowchart of the emergency notification apparatus according to an embodiment of the present invention,
4 is a view showing an example of determining the length between the eyes, the nose, the ear and the mouth in the face of the subject of the video signal in order to identify a specific sensing object (for example, an employee)
5 is an example of a table referred to in order to determine an emergency situation.

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

An example of an overall security system including an emergency notification control apparatus according to an embodiment of the present invention is shown in FIG.

As shown in the figure, the entire security system includes at least one surveillance camera 200, an emergency notification device 100, and a monitoring server 300.

Here, the surveillance camera 200 performs a function of capturing an external subject, and can be distinguished into a CMOS type or a CCD type according to the type of the imaging device.

Such a surveillance camera 200 can be installed in various places where crime prevention is required such as a home, a shopping mall, a store, etc. In this embodiment, the surveillance camera 200 is installed in a convenience store.

The surveillance camera 200 photographs a subject, that is, a convenience store clerk or a customer at a convenience store, and transmits a video signal corresponding to the captured video image, since it corresponds to a known technology, and thus a detailed description thereof will be omitted.

 The monitoring server 300 is a predetermined server on the network that performs a function of receiving a warning alarm signal generated in the emergency situation notification device 100 as described later. Also, the monitoring server 300 may receive the video signal photographed by the surveillance camera 200 via the emergency situation notification device 100 as will be described later. A detailed description of the process of receiving the photographed video signal will be given later.

The monitoring server 300 may be a server managed by a private security company, or a server managed in association with a public institution.

The emergency notification device 100 is connected to the surveillance camera 200 to analyze the image of the subject photographed by the surveillance camera 200 and generate a warning alarm signal according to the situation.

The detailed functional blocks of the emergency notification device are as shown in FIG.

As shown in the figure, the emergency notification device may include a database unit 110, a determination unit 120, a selection unit 130, a warning alarm unit 140, and a video signal receiving unit 150.

First, the video signal receiving unit 150 performs a function of receiving a video signal photographed by the surveillance camera 200. The video signal receiving unit 150 and the surveillance camera 200 may be connected to each other via a coaxial cable or a 4P multi cable, and the connection configuration may be various. The connection configuration with the surveillance camera 200 itself Is a known technology, and thus a detailed description thereof will be omitted.

The database unit 110 includes a table in which the detection subject information and the behavior pattern information are matched.

Here, the detection subject information is basic information for selecting the detection subject, as described later, and may correspond to, for example, information about a face feature of a specific person, that is, a convenience store employee. As an example of information about the facial feature, it may be information about the distance between at least two of the main organs attached to the face of a person, such as the eyes, nose, mouth, and ears.

In addition, the information to be detected may be information on the outline of the entire face, size, hairstyle, whether or not the hat is worn, the color of the wearing hat, whether the uniform is worn, uniform color,

As described later, the behavior pattern information corresponds to a specific action performed by a specific person. For example, the behavior pattern information can be represented by a person's actions such as " an action to shake a specific arm, It is information about specific behavior.

That is, at least one piece of sensing object information and behavior pattern information are matched to the database unit 110. For example, the characteristic information of the convenience store clerk (that is, sensing target information) Information about the movement to be operated (that is, the action pattern information) may be matched, or the face characteristic information (corresponding to the detection target information) of a person who writes a mask on the face such as robbery and the knife of the robber (Corresponding to behavior pattern information) of a specific action such as a threat or the like may be matched.

A more detailed description of the detection subject information and the behavior pattern information will be given later.

As another example, time information may be further matched to the database. That is, the database 110 may be configured such that the detection subject information, the behavior pattern information, and the time information are matched with each other. Here, the time information may be a specific time or a specific time range.

The determination unit 120 analyzes the image of the subject photographed through the surveillance camera 200 and determines the shape and behavior pattern of each subject.

That is, the determination unit 120 analyzes the image of the subject included in the video signal received by the video signal receiving unit 150, and determines the shape of each subject and the behavior pattern corresponding to the motion of the subject.

The shape information of the subject determined by the determination unit 120 may include the shape of the eyes, nose, mouth, ear, and distance information of the face feature, if the subject is a human, as mentioned in the detection subject information .

That is, the judging unit 120 distinguishes the face part and the body part of each object (for example, the convenience store clerk and the guest) included in the image photographed through the surveillance camera 200 on the basis of the outline of the subject, After determining the position of the eyes, nose, mouth, and ears in the part, the distance between at least two of the eyes, nose, mouth, and ear can be calculated.

In addition, information on the shape of the subject may include information on the outline, size, hairstyle, wearing of hat, color of wearing hat, wearing uniform, color of uniform, wearing mask.

The determination unit 120 also analyzes the image of the subject photographed through the surveillance camera 200 and determines a change in illuminance. For this, though not shown in the drawing, the determination unit 120 may be configured to include an illumination sensor.

The selecting unit 130 compares the shape of each of the objects determined by the determining unit 120 with the subject information to select a subject to be sensed among the subjects. For example, when the face feature of the convenience store clerk is included in the database unit 110 and the face feature of the subject determined by the determination unit 120 is the same as the face feature of the convenience store clerk, As a detection target.

For example, the selection unit 130 compares the distance between at least two of the eyes, the nose, the mouth, and the ear calculated by the determination unit 120 with the sensing target information stored in the database unit 110, You can do it.

Since the sensing target selected by the selecting unit 130 is not limited to one, all of the plurality of subjects may be selected as sensing targets.

The warning alarm unit 140 generates a warning alarm signal when the behavior pattern of the sensing object selected by the selection unit 130 matches the behavior pattern information matched in the database.

For example, the warning alarm unit 140 generates a warning alarm signal when the convenience store clerk selected as the sensing target in the selection unit 130 performs a specific action registered in the database unit 110 corresponding to the convenience store clerk , Or when a robber selected as a detection target performs a specific action registered in the database unit 110 in response to the robber.

For example, the face feature (size and interval of eyes, nose, mouth, ear, etc.) of the convenience store clerk A and the face feature of the robbery B (wearing a mask) are registered in the database unit 110, The facial features of clerk A are to move both arms to the side of the ear and shake it to the left and right at the same time. In the face feature of the robber B, the knife is used to threaten another object with a knife. When the shaking motion is matched, the warning alarm unit 140 moves the arms of the convenience store A to the side of the ear and shakes it simultaneously to the left and right, while the subject convenience store point A on the image is selected as the object to be sensed A warning alarm signal is generated.

Similarly, the warning alarm unit 140 generates a warning alarm signal when the subject striking person B on the image image shakes the hand with the knife back and forth, right and left.

On the other hand, in the case where the object A corresponding to the convenience store clerk performs the operation of swinging the hand with the knife back and forth, left and right, or the case where the B object corresponding to the robbery raises both arms to the side of the ear and shakes the left and right simultaneously Warning Do not generate an alarm signal.

That is, the warning alarm unit 140 generates a warning alarm signal only when the subject selected as the sensing target operates a specific behavior pattern matching the sensing target.

This is different from generating a warning alarm signal when an arbitrary object of the image displayed on the image implements a specific behavior pattern, or when a subject having a specific form is found in the image (for example, And when a subject is found).

When a certain object generates a warning alarm signal whenever a specific behavior pattern defined in the database unit 110 is operated or when a subject having a specific form is found in the image image, The possibility of error is very high.

That is, even though the actual situation is not an emergency situation, the emergency situation notification device 100 erroneously determines that it is an emergency situation.

On the other hand, when a specific object to be sensed performs a specific behavior pattern matched to the object to be sensed, generating a warning alarm signal reduces the possibility of erroneously determining the emergency situation.

Here, the warning alarm unit 140, which has generated the warning alarm signal, may operate the warning light and the siren directly connected thereto using the warning alarm signal, or may transmit the generated warning alarm signal to the external monitoring server 300 ).

In particular, when the alarming alarm signal is transmitted to the external monitoring server 300, the alarming alarm unit 140 may transmit the video signal received from the monitoring camera 200 to the external monitoring server 300. Of course, the warning alarm unit 140 may always transmit the video signal received from the surveillance camera 200 to the monitoring server 300, but may transmit the video signal to the monitoring server 300 only when it is determined that an emergency situation has occurred It is possible.

Meanwhile, the warning alarm unit 140 may further generate the warning alarm signal using the time information stored in the database unit 110. [

That is, the warning alarm unit 140 may generate a warning alarm signal when the behavior pattern of the sensing object selected by the selection unit 130 and the current time match the behavior pattern information and the time information matched in the database.

For example, even if the convenience store clerk performs the same action, it can generate a warning alarm signal only at night without generating a warning alarm signal during the daytime. In other words, if a specific behavior pattern of a convenience store clerk occurs on a day-to-day basis, but does not occur routinely at night, if an alarm signal is generated without considering the time when such a behavior pattern occurs, Warning A problem arises that an alarm signal is generated.

This can deceive the robber by acting in the daytime routine behavior pattern when the night robber breaks in. That is, the robber can not be aware of the operation of the convenience store clerk itself so that a warning alarm signal is generated.

Meanwhile, the warning alarm unit 140 may generate a warning alarm signal even when the variation of the illuminance determined by the determination unit 120 exceeds a predetermined range.

For example, when the robber can cut off the power supplied to the convenience store, the warning alarm unit 140 detects a sudden change in the reference illuminance and generates a warning alarm signal. Time information may be taken into account for the degree of occurrence of the warning alarm signal in accordance with the change in illuminance.

For example, the warning alarm unit 140 generates a warning alarm signal only when the brightness by the electric lamp suddenly changes at night without generating a warning alarm signal when the weather suddenly becomes dark due to the dimming of the day.

Hereinafter, a control flow of the emergency notification apparatus 100 according to an embodiment of the present invention will be described with reference to FIG.

The emergency situation notification device 100 recognizes the face of the subject included in the video signal received from the surveillance camera 200 and identifies the employee (step S1).

In order to identify the employee, the emergency notification apparatus 100 may compare the interval between at least two of the eyes, nose, mouth, and ear of the subject's face with pre-registered information as shown in FIG.

Next, the emergency notification apparatus 100 determines whether the current time is within a predetermined time range (step S3). If the current time is within the preset time range, ).

If the behavior pattern of the determined employee corresponds to the previously registered behavior pattern (step S7), the emergency situation notification device 100 generates and sends a warning alarm message (step S9).

FIG. 5 shows an example of a table used in the process of determining the behavior pattern of the employee by the emergency situation notification device 100. FIG.

The emergency situation notification apparatus 100 detects the occurrence of a specific object on the left and right sides of the face of the employee through the analysis of the object, and when the object occurrence time is simultaneously maintained for three seconds on the left and right faces of the face, And detects an emergency situation. At this time, it goes without saying that the emergency notification apparatus 100 generates a warning alarm message.

As another example, when the movement of the face of the employee occurs, it is detected, and when the operation of moving the face to the left or to the right occurs repeatedly, it is determined that the employee is in an emergency and may judge that the situation is an emergency.

Needless to say, the process of performing each of the above-described embodiments can be performed by a program stored in a predetermined recording medium (for example, a computer readable medium).

The present invention is not limited to the above-described specific embodiments, and various modifications and changes may be made without departing from the gist of the present invention. It is to be understood that such variations and modifications are intended to be included in the scope of the appended claims.

100: Emergency notification device 200: Surveillance camera
300: Monitoring server 110: Database part
120: Judgment section 130:
140: warning alarm unit 150: video signal receiving unit

Claims (9)

delete 1. An emergency situation notification device connected to a surveillance camera,
A database unit in which detection target information and behavior pattern information are matched;
A determination unit for analyzing an image of a subject photographed through the surveillance camera and determining a shape and a behavior pattern of each subject;
A selection unit configured to compare a shape of each of the objects determined by the determination unit with the detection subject information to select a subject to be sensed;
And a warning alarm unit for generating a warning alarm signal when the behavior pattern of the sensing object selected by the selection unit matches the behavior pattern information matched in the database,
Wherein the database stores the detection subject information, the behavior pattern information, and the time information,
Wherein the warning alarm unit generates a warning alarm signal when the behavior pattern of the sensing object selected by the selection unit and the current time match the behavior pattern information and the time information matched in the database. 1. An emergency situation notification device connected to a surveillance camera,
A database unit in which detection target information and behavior pattern information are matched;
A determination unit for analyzing an image of a subject photographed through the surveillance camera and determining a shape and a behavior pattern of each subject;
A selection unit configured to compare a shape of each of the objects determined by the determination unit with the detection subject information to select a subject to be sensed;
And a warning alarm unit for generating a warning alarm signal when the behavior pattern of the sensing object selected by the selection unit matches the behavior pattern information matched in the database,
Wherein the determination unit analyzes an image of a subject photographed through the surveillance camera to determine a change in illuminance,
Wherein the warning alarm unit generates a warning alarm signal when the variation of the illuminance determined by the determination unit exceeds a predetermined range. The method according to claim 2 or 3,
Wherein the detection subject information corresponds to a distance between at least two of the eyes, nose, mouth, and ear of a person,
In the determination of the shape of the subject, the determination unit distinguishes the face portion and the body portion of each subject based on the outline of the subject, determines the positions of the eyes, nose, mouth, and ears in the distinguished face portion, , The distance between at least two of the ears is calculated,
Wherein the selecting unit selects the sensing target by comparing the sensing target information with a distance between at least two of the eyes, nose, mouth, and ear calculated by the determining unit. delete A control method of an emergency situation notification device having a database matching detection target information and behavior pattern information and connected to a surveillance camera,
(a) analyzing an image of a subject photographed through the surveillance camera to determine the shape and behavior pattern of each subject;
(b) comparing a type of each of the objects determined in the step (a) with the detection subject information to select a subject to be sensed;
(c) generating a warning alarm signal when the behavior pattern of the sensing object selected in the step (b) matches the behavior pattern information matched in the database,
Wherein the database stores the detection subject information, the behavior pattern information, and the time information,
Wherein in the step (c), a warning alarm signal is generated when the behavior pattern and the current time of the sensing object selected in the step (b) match the behavior pattern information and the time information matched in the database Control method of notification device. A control method of an emergency situation notification device having a database matching detection target information and behavior pattern information and connected to a surveillance camera,
(a) analyzing an image of a subject photographed through the surveillance camera to determine the shape and behavior pattern of each subject;
(b) comparing a type of each of the objects determined in the step (a) with the detection subject information to select a subject to be sensed;
(c) generating a warning alarm signal when the behavior pattern of the sensing object selected in the step (b) matches the behavior pattern information matched in the database,
Analyzing an image of a subject photographed through the surveillance camera to determine a change in illuminance;
Further comprising the step of generating a warning alarm signal when the change in illuminance exceeds a predetermined range. 8. The method according to claim 6 or 7,
Wherein the detection subject information corresponds to a distance between at least two of the eyes, nose, mouth, and ear of a person,
The step of determining the shape of the subject in the step (a) includes the steps of: distinguishing the face part and the body part of each subject based on the outline of the subject; Calculating a distance between at least two of the eyes, nose, mouth, and ear after determining the positions of the eyes, nose, mouth, and ears in the distinguished face part,
Wherein in the step (b), the sensing object is selected by comparing the sensing object information with a distance between at least two of the eyes, nose, mouth and ear calculated in the step (a) / RTI > A computer-readable recording medium recording a program for executing the method of claim 6 or 7.

Images