KR20170010426A - Method for alarming a danger using the camera - Google Patents

Abstract

The present invention relates to a method for warning of danger using a camera, and more particularly to a method for warning of danger using a camera, which detects a motion of an object directly from video acquired via a camera, which determines a situation in which the motion of the object stops for a time equal to or longer than a predetermined time to be a dangerous situation, and which transmits a danger signal, thereby enabling a probability of a dangerous situation to be immediately detected even when a dangerous element is not detected from an actual background. For this purpose, according to an aspect of the present invention, a method for warning of danger using a camera, which is performed by a system for warning of danger configured to include a control unit adapted to detect danger of a person disposed in video by using the video acquired via a camera photographing a specific place and configured to issue warning, comprises: an object detection step of detecting an object to be monitored from the video; a motion detection step of detecting a motion of the object; and a danger warning step of transmitting a danger signal when the motion of the object stops for a time equal to or longer than a predetermined time.

Description

[0001] The present invention relates to a method for alarming a camera using a camera,

More particularly, the present invention relates to a danger warning method using a camera, and more particularly, to detecting a motion of an object directly from a camera image, determining that a motion of the object has stopped for a predetermined time or longer, The present invention relates to a risk warning method using a camera that can immediately detect the possibility of a dangerous situation even if the risk is not detected.

Recently, the installation of surveillance cameras such as CCTV is rapidly increasing, and various technologies are being developed to enhance the surveillance function through such CCTV cameras. There are also many social support measures for elderly people and those who are uncomfortable with the aging of society. Especially, elderly people who live alone need to be careful about this because they often happen like fall.

The fall of the elderly is considered as a major social problem in recent years as the elderly population is rapidly increasing due to a serious problem in health and life. Elderly people are more likely to fall into a dangerous situation, and because of their inability to cope with dangerous situations themselves, serious problems can arise if proper measures are not taken. Therefore, it is necessary to take prompt action and countermeasures in case of danger such as falling.

This also applies to young children and workers in construction workplaces. It is more likely to fall into a dangerous situation than an ordinary adult, and it is important to take immediate action to recognize it immediately when facing such a dangerous situation.

For this purpose, many techniques for detecting dangerous behavior using various sensors have been proposed. For example, Korean Patent No. 10-1069001 discloses a technique for detecting falling by comparing a sensor pattern with a standard pattern by attaching a plurality of sensors to the floor (Korean Patent No. 10-1069001) and attaching an accelerometer and a gyroscope sensor to the object, (Korean Patent No. 10-1196296) is disclosed.

However, in such a case, it is necessary to install a sensor at every place where dangerous behavior such as a fall is expected, so that it is costly and troublesome to attach the sensor to the body. Especially, it is difficult to interpret data for fall detection .

In addition, since it is a method of predicting the possibility of danger by recognizing the environment of the surrounding background image when the sensor method is not directly attached to the body of the target object, the motion of the target object, It is judged as a natural state, and when there is a problem in the body function due to a chronic illness or the like, there is a problem that it can not be detected at all through the image.

SUMMARY OF THE INVENTION The present invention has been made in order to solve the above problems, and it is an object of the present invention to provide a method and apparatus for detecting a motion of an object directly from a camera image, Thereby providing a risk warning method using a camera that can detect the possibility of a dangerous situation immediately even if a dangerous factor on the actual background is not detected.

Another object of the present invention is to provide a method and apparatus for detecting a motion of an object when the motion of the object has been stopped for a predetermined time or longer before the motion change per unit time of the object has occurred more than a threshold value, When a motion change of an object occurs according to a dangerous motion pattern defined to correspond to a feature element of the detected object, it is determined that the object is in a high risk state, thereby minimizing the possibility of a serious situation being recognized And to provide a danger warning method using a camera.

According to an aspect of the present invention, there is provided a risk warning method performed by a risk warning system in which a control unit detects a risk of a person in an image using an image of a camera capturing a specific place, An object detecting step of detecting a monitored object in the image; A motion detection step of detecting motion of the object; And a danger warning step of sending a danger signal when motion of the object is not detected for a predetermined time or more.

And a high-risk alarming step of sending a high-risk signal when the motion of the object is not detected for a predetermined time or more after a motion change per unit time of the object occurs at a threshold value or more. And a high risk alarming step of sending a high risk signal when the motion of the object coincides with a predetermined risk pattern and the motion of the object is not detected for a predetermined time or longer after the motion change occurs.

Here, the object detection step may include grouping when a plurality of objects are detected, and determining a next position to be photographed by the camera through a motion change of the object, and the motion detection step may detect motion of each group It is possible. At this time, the object detecting step removes noise based on the grouping, and determines a next position to be photographed by the camera by obtaining a center of gravity by giving a weight according to the size of the motion to each group.

In addition, the object detection step may detect the feature element of the object, and the motion detection step may detect the motion of the object according to the dangerous motion pattern defined to correspond to the feature element, And a high risk warning step of transmitting a high risk signal when the motion of the object is not detected for a predetermined time or longer after the motion change occurs.

According to the present invention, when a motion of an object is detected for a predetermined time or longer by detecting motion of an object directly from a camera image, it is determined that the motion of the object is a dangerous situation and a danger signal is transmitted. The reliability of the detection of the dangerous situation can be improved. According to the present invention, the elderly living alone can accurately detect whether or not a dangerous behavior such as a fall occurs. Further, according to the present invention, it is possible to notify the external related organizations in real time about the detection of the dangerous situation, thereby enabling urgent measures in the related institutions.

1 is a block diagram of a risk warning system using a camera according to an embodiment of the present invention.
2 is a flowchart of a risk warning method using a camera according to the first embodiment of the present invention.
3 is a flowchart of a risk warning method using a camera according to the second embodiment of the present invention.
4 is a flowchart of a risk warning method using a camera according to the third embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described in detail with reference to the accompanying drawings. It is to be noted that like elements in the drawings are denoted by the same reference numerals whenever possible. It should be understood, however, that the terminology or words of the present specification and claims should not be construed in an ordinary sense or in a dictionary, and that the inventors shall not be limited to the concept of a term It should be construed in accordance with the meaning and concept consistent with the technical idea of the present invention based on the principle that it can be properly defined. Therefore, the embodiments described in the present specification and the configurations shown in the drawings are merely the most preferred embodiments of the present invention, and not all of the technical ideas of the present invention are described. Therefore, It is to be understood that equivalents and modifications are possible.

1 is a block diagram of a risk warning system using a camera according to an embodiment of the present invention.

Referring to FIG. 1, a camera-based critical alarm system 100 according to an exemplary embodiment of the present invention includes a camera 110, a client device 120, a communication network 130, and a timer 140.

The camera 110 acquires an image of the surrounding environment and processes image data according to the image. And acquires the image of the object in real time to grasp the motion of the target object. And an infrared camera capable of identifying an object at night.

The client device 120 detects an object using the image acquired by the camera 110 and detects motion of the object to determine whether or not the object is in a dangerous state and warns it. The client device 120 includes an object detection module 121, a motion detection module 122, a controller 123, and a danger warning module 125 as shown in the figure. In another embodiment, the client device 120 may further include at least one or more of a display unit 124, a memory unit 126, and a communication unit 127.

The object detection module 121 extracts a color image and a depth image from the peripheral image obtained by the camera 110. [ HSV is a model that assigns a specific color to hue, saturation and brightness. It converts all colors to RGB (Red, Green, , And Blue), the color is represented by a combination of color (H), saturation (S), and brightness (V), unlike a known RGB color model.

In addition, the object detection module 121 detects an object using, for example, an enhanced double frame difference method from the background image generated as described above. Specifically, by applying the improved double frame difference method, the color difference of each pixel between the background image and the extracted color image is detected for a predetermined time, and the object is detected using the pixel whose detected color difference is equal to or greater than a preset threshold value do.

The motion detection module 122 obtains moment analysis information from the detected object, extracts feature information of the object using the obtained moment analysis information and depth information of the object according to the depth image, And tracks the motion of the object from the feature information of the object. In the motion detection of such an object, the position of the center point of the detected object is calculated, the search window is set based on the position of the center point of the detected object, and the center position of the object detected in the next frame is calculated, If it is in the window, keep track of the movement of the object. The center point may be, for example, the center of gravity of the object or may be the physical center of the location of the object represented by the color image.

The control unit 123 controls the overall operation of the client apparatus 120. [ In this embodiment, the control unit 123 determines whether or not the dangerous situation of the corresponding object occurs according to whether the movement of the object detected from the motion detection module 122 is stopped for a predetermined time or longer. It is determined that there is a possibility that a problem has occurred in the personal image if the motion of the object is not detected over the set time through the timer 140. [

The danger warning module 125 sends a danger signal when the controller 123 determines that the state of the target object is dangerous. Also, the control unit 123 may determine that the situation is more urgent according to the passage of time or when a specific condition is established, and may transmit the high-risk signal from the danger warning module 125. As such means, an LED lighting device capable of blinking, an acoustic generator capable of sending beep sound or sirens, and the like can be used.

As described above, the risk warning system 100 using the camera according to the present invention may further include at least one or more of a display unit 124, a memory unit 126, and a communication unit 127. The display unit 124 displays a dangerous situation detection message when the control unit 123 determines that the object is in a dangerous situation. Accordingly, in the present invention, the display unit 124 and the danger warning module 125 can be informed to each other in a visual and auditory manner. The memory unit 126 stores various data and information. For example, a dangerous motion pattern that is defined to correspond to a feature element of each object, a dangerous motion pattern that appears when the person is at risk, and the like. The communication unit 127 performs communication with the external device by the control signal of the control unit 123. [

The communication network 130 is responsible for communication between the client device 120 and a plurality of external institutional servers 200. The communication network 130 may include a wired / wireless Internet network. The external institution server 200 receives various kinds of information transmitted from the client device 120. [ Particularly, when the control unit 123 detects a dangerous situation, it transmits a dangerous situation detection message to at least one external institutional server 200, and the external institutional server 200 receives the dangerous situation detection message, The situation can be grasped. Here, the external institution server 200 is a server apparatus provided in at least one external organization for managing and managing the dangerous behavior and the emergency situation, for example, a server apparatus provided in a police station, a fire station, a hospital, an insurance company or the like.

2 is a flowchart of a risk warning method using a camera according to the first embodiment of the present invention.

Referring to FIG. 2, the danger warning method performed by the risk warning system 100, in which the controller 123 detects and warns the user of the risk of a person in the image using the camera image, includes an object detection step S10, (S20) and a danger warning step (S30).

The object detection step (S10) detects the object to be monitored in the camera image through the object detection module (121). The object to be monitored is mostly a person, and any person who needs to know and monitor the danger immediately through the camera can be the target of the user. Accordingly, it is preferable that the camera is installed by designating the position so that the active area of the target object to be monitored can be photographed.

The motion detection step S20 detects the motion of the object. The motion of an object has various motions according to the situation, and this ordinary motion is detected in the camera image. The motion of the object is collected through the motion detection module 122 by information such as the size and direction of the motion, the activity area, and the like.

Meanwhile, the object detection step S10 may determine the next position to be photographed by the camera through the motion change of the object. Since the motion of an object can generally predict its direction, it is possible to move the photographing direction of a mobile or rotary camera together according to this prediction. Since the object to be monitored does not remain in one place but can move over a wider area, a change in the camera's photographing position due to such a motion change helps to more stably monitor the object.

In a case where a plurality of objects are simultaneously detected in the object detecting step S10, it is possible to group them. In the object detecting step S10, noise is removed based on the grouping, and for each group, By assigning weights to determine the center of gravity, the camera can determine the next position to take.

If a background other than the grouped object group is judged as noise and removed, only the images of the object group are left. In the noise-removed image, the size of the motion is determined for each group, Depending on the motion change, the camera can determine the next position to take. According to this method, even when a plurality of non-single objects are simultaneously detected in the camera image, it is possible to continuously track and monitor the main object so as not to deviate from the area of the image.

The danger warning step S30 is a step of sending a danger signal when the motion of the object is not detected for a predetermined time or more. During the activity, the observed object continues to show unspecific or specific motion. If there is no change in the motion for a certain period of time after the motion is stopped at some point, there is a possibility that the problem of the object has occurred. If a problem occurs in the image of the elderly, the child, the worker, etc., it is necessary to identify and take measures as soon as possible. For this reason, if the motion of the object is not detected for a predetermined time, the danger warning module 125 immediately sends out the danger signal.

2, in the risk warning method using the camera according to the first embodiment of the present invention, in addition to the object detection step S10, the motion detection step S20 and the danger warning step S30, Step S31 may be included.

If the motion of the object is not detected for a predetermined time or longer, if it is a danger warning step S30, then the high risk alarm step S31 is a step of, after the motion change per unit time of the object occurs above the threshold, If it is not detected for a predetermined time or longer, a high-risk signal is transmitted.

The fact that the motion change per unit time of the object occurred above the threshold value can mean that a large event occurred instantaneously in the object's identity. For example, if an event occurs, such as skidding or an explosion at a construction work site, the motion of the target object changes instantaneously. Since a large motion is generated in a short time, it is possible to predict the change of the object by detecting that the motion change per unit time occurs above a predetermined threshold value.

However, even if the motion change per unit time is great, it is correct to judge that the object is not a dangerous situation if ordinary motion is detected again from immediately after that. Since there are so many situations in which a large movement can be taken for a while during movement or activity, the momentary change in motion per unit time can not be determined as a dangerous situation.

Therefore, in the first embodiment of the present invention, the control unit 123 determines that the motion of the object is not dangerous for a predetermined time or longer after the motion change per unit time of the object is generated above the threshold, Module 125 to transmit the high-risk signal. Such a high risk situation is generally more likely to require more dangerous and urgent action than when the motion of the object is not detected for a certain period of time. Accordingly, the risk alert module 125 may provide a stronger notification High-risk signal.

The more dangerous the situation, the more dangerous it is, the stronger the alarm should be. The alarm system according to the embodiment of the present invention can be as shown in the following table.

Risk Level
Alarm means Risk situation High risk situation 3 minutes 5 minutes LED light on Flashing Fast flashing Alarm sound - Beep sound Alarm sound (siren)

3 is a flowchart of a risk warning method using a camera according to the second embodiment of the present invention.

Referring to FIG. 3, in the second embodiment of the present invention, the high risk warning step S31 is performed in a manner different from the first embodiment. A high risk signal is transmitted when a motion change of an object occurs first in accordance with a predetermined risk pattern and after the motion change is not detected for a predetermined time or longer.

When a plurality of dangerous pattern motions are stored in the memory unit 126 and motion of the same pattern as the previously stored dangerous patterns is detected during the continuous detection of the motion change of the object, It can mean that a situation has occurred. For example, when slipping and falling, there is a series of motion patterns that stretch or turn hands up or down, the height of the upper body is instantaneously lowered, and the lower body collapses. If this same motion appears in the object, Can be estimated.

Therefore, in the second embodiment of the present invention, it is determined that the motion of the object is very dangerous when the motion of the object is not detected for a predetermined time or longer after the motion change of the object is made to coincide with the predetermined risk pattern, and the high risk signal is transmitted. The method of transmitting the high-risk signal is the same as that of the first embodiment.

4 is a flowchart of a risk warning method using a camera according to the third embodiment of the present invention.

Referring to FIG. 4, in the third embodiment of the present invention, a feature element of an object is detected in an object detection step (S10). For example, it detects a walking aid such as a hair color or a bent waist or a walking stick to detect that it is an elderly person with a staff, or detects a young child through a small key.

If the feature element of the object is detected in the object detection step (S10), the motion detection step (S20) can detect whether or not the object takes the same motion according to the dangerous motion pattern defined to correspond to the feature element . When the dangerous motion pattern stored in the memory unit 126 is displayed so as to correspond to a feature element of the object such as a sudden break of a cane or a grabbing of a forehead or the like, You can predict the change of the object's identity.

A high risk warning step S31 for sending a high risk signal when the motion of the object is not detected for a predetermined time or longer after the motion change of the object first occurs so as to match with the dangerous motion pattern corresponding to the characteristic element of the object. The manner of transmitting the high-risk signal is the same as that of the first embodiment and the second embodiment.

Although the present invention has been described in connection with the above-mentioned preferred embodiments, it is possible to make various modifications and variations without departing from the spirit and scope of the invention. Accordingly, the scope of the appended claims should include all such modifications and changes as fall within the scope of the present invention.

100: Critical alarm system
110: camera
120: Client device
121: Object detection module 122: Motion detection module
123: control unit 124:
125: DANGER ALERT MODULE 126: MEMORY
127:
130: communication network 140: timer
200: external agency server

Claims (3)

A risk warning method performed by a risk warning system in which a control unit detects and alerts the danger of a person in an image using a camera image,
An object detecting step of detecting one monitoring object in the image; A motion detection step of detecting motion of the object; And a danger warning step of sending a danger signal when the motion of the object is not detected for a predetermined time or more,
A motion change per unit time of the one object occurs at a threshold value or more,
The motion change of the one object coincides with the danger pattern of one object preset in the memory unit,
Detecting a characteristic element of the one object in the object detection step and detecting a motion of the one object in the motion detection step according to the dangerous motion pattern defined to correspond to the characteristic element in the motion detection step, A high risk alarm step of transmitting a high risk signal when the motion of the object is not detected for a predetermined time or more after the motion change occurs;
Wherein the camera is provided with a camera.
The method according to claim 1,
Wherein the object detection step includes the step of determining a next position to be photographed by the camera through a motion change of the object,
Wherein the motion detection step detects motion for each group.
3. The method of claim 2,
Wherein the object detecting step removes noise based on the grouping and determines a next position to be photographed by the camera by calculating a center of gravity by assigning a weight according to the size of the motion to each group. Risk warning method.

Images