KR20130122409A - Apparatus for detecting of single elderly persons' abnormal situation using image and method thereof - Google Patents

Abstract

In order to detect an abnormal situation, videos taken in real time are obtained from video cameras installed in multiple areas within a house, the number of variation values between former and latter frames of the obtained videos according to pixels is calculated, the number of pixels in which a calculated variation value is over a predetermined variation value is detected, an activity amount of an object is calculated by accumulating the number of the pixels detected within a predetermined time range, the calculated activity amount is stored corresponding to previously classified date and time slot, the value of the calculated activity amount and the value of a previously stored activity amount are compared, occurrence of an abnormal situation is determined based on a difference value between the compared activity amounts, and an activity amount which is in the same time slot as a currently calculated activity amount among previously stored activity amounts is detected and compared. [Reference numerals] (AA) Start;(BB) End;(S310) Obtain images of multiple areas;(S320) Calculate pixel variation values between the frames of the images;(S330) Calculate activity values;(S340) Determine an abnormal situation by comparing the activity values within a predetermined reference period;(S350) Notify about the abnormal situation

Description

Apparatus and method for detecting abnormal occurrence of elderly living alone using image data {APPARATUS FOR DETECTING OF SINGLE ELDERLY PERSONS ’ABNORMAL SITUATION USING IMAGE AND METHOD THEREOF}

The present invention relates to an abnormal situation detection apparatus and method for detecting the occurrence of an abnormal situation for the elderly living alone using an image photographed through a photographing device installed in the living space of the elderly living alone.

Recently, as the elderly population is rapidly increasing, new research fields for improving the welfare of the elderly by applying IT technology are actively progressing.

Research trends in these related areas have shown that the elderly's movements can be closely analyzed, starting with a relatively simple manual monitoring system that allows them to visually monitor the elderly living alone in the home and alert the relevant parties when unusual signs of abnormality are detected. Research has been conducted in various fields, ranging from early detection of abnormal signs and systems to efficiently determine abnormal signs.

On the other hand, in the existing elderly living monitoring method, a method that can determine whether the abnormality occurred in the elderly living alone using information obtained from various sensors installed in the house has been proposed. In such an existing abnormality determination method, a method of utilizing data acquired from a sensor installed in a house through a sequence alignment method and a motion vector method has been proposed.

In this regard, Korean Patent Laid-Open Publication No. 10-2008-0026326 (A device and method for monitoring the elderly motion image using the principal component method) analyzes the elderly motion image using the principal component method that analyzes the elderly's motion to determine an emergency situation. An apparatus and method are disclosed. Specifically, Korean Laid-Open Patent Publication No. 10-2008-0026326 discloses an image preprocessing process that separates a silhouette of a person from a background of an image and outputs a binarized image from a video signal captured by a surveillance camera and converted into a digital signal. In addition, PCA algorithm is applied to analyze the posture change by using the change of proportions of the eigenvectors of the human silhouette as a parameter by applying the PCA algorithm to the information of the separated human silhouette during image preprocessing. If a sudden change in the ratio of the eigenvectors of the silhouette of the person is detected, it is determined whether it is an emergency situation. have.

The present invention is to provide an apparatus and method for detecting the occurrence of abnormal situations for the elderly living alone using a photographing device installed in a residential environment.

Abnormal situation detection apparatus according to an aspect of the present invention for achieving the above technical problem, a plurality of photographing unit is installed in a plurality of zones in the house to take an image; Computing the difference of each pixel value between the front and rear frame of the captured image to detect the number of pixels having a deviation or more than a predetermined reference deviation, and to calculate the amount of activity of the object by adding the number of the detected pixels within a predetermined time range Activity calculation unit; An activity data storage unit classifying the calculated activity amount by date and time zone and storing the calculated activity amount as activity amount data; And an abnormal situation determination unit determining an occurrence of an abnormal situation in real time based on the calculated activity amount and the previously stored activity amount data, wherein the abnormal situation determination unit is configured to determine the same amount of activity amount as that of another date among the previously stored activity amount data. A value is compared with a value of the calculated activity amount, and the occurrence of the abnormal situation is determined based on the difference value between the compared activity amounts.

The abnormal situation detection method using the abnormal situation detection apparatus according to another aspect of the present invention includes the steps of: (a) acquiring an image captured in real time from a photographing apparatus installed in a plurality of zones in a house; calculating a change value of each pixel value between front and rear frames of the obtained image; (c) detecting the number of pixels whose calculated change value is equal to or greater than a predetermined deviation value; calculating an amount of activity of an object by accumulating the number of pixels detected within a preset time range; (e) storing the calculated amount of activity to correspond to a previously classified date and time zone; (f) comparing the value of the previously stored activity amount with the value of the calculated activity amount; And (g) determining an occurrence of an abnormal situation based on a difference value between the compared activity amounts, wherein step (f) detects the calculated activity amount and activity amount in the same time period among the previously stored activity amounts. The comparison is made.

According to any one of the above-described problem solving means of the present invention, by detecting the amount of activity of the elderly living alone in real time by using the images taken through a plurality of photographing unit installed in a house where the elderly living alone, the abnormal situation caused to the elderly living alone Can be detected quickly.

In addition, according to any one of the problem solving means of the present invention, by detecting the amount of movement of the object in the image based on the amount of change between pixels in the frame in the image, there is an effect that it is possible to simply and quickly calculate the amount of activity of the elderly living alone. .

In addition, according to any one of the problem solving means of the present invention, when it is determined that an abnormal situation occurs for the elderly living alone, the activity amount data corresponding to the current time and the activity amount data corresponding to the same time period among the activity amount data previously stored for each date and time By judging the occurrence of an abnormal situation, it is possible to quickly detect the condition of the elderly different from the other days, that is, in general, in consideration of the characteristics of life in which the elderly lives in a relatively same pattern, it is possible to set a preset condition even when the elderly are alone. Compared to other days within the period, it is possible to efficiently determine when activity is low or abnormal.

In addition, according to any one of the problem solving means of the present invention, to identify the number of people in the living space in real time by identifying the access of the person on the basis of the image taken from the photographing unit installed in the entrance area of the elderly living alone In other words, if there is only one number of people in the family, the elderly are determined to be alone (or occupied).

1 is a block diagram showing the configuration of an abnormal situation detection apparatus according to an embodiment of the present invention.
2 is a view for explaining an example of an activity data storage method according to an embodiment of the present invention.
3 is a flowchart illustrating an abnormal situation detection method according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings, which will be readily apparent to those skilled in the art. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. In order to clearly illustrate the present invention, parts not related to the description are omitted, and similar parts are denoted by like reference characters throughout the specification.

Throughout the specification, when a part is referred to as being "connected" to another part, it includes not only "directly connected" but also "electrically connected" with another part in between . Also, when an element is referred to as "comprising ", it means that it can include other elements as well, without departing from the other elements unless specifically stated otherwise.

1 is a block diagram showing the configuration of an abnormal situation detection apparatus according to an embodiment of the present invention.

As shown in FIG. 1, the abnormal situation detecting apparatus 100 according to an exemplary embodiment may include a plurality of photographing units 111, 112, 113, an image processing unit 120, an occupant calculation unit 130, Activity amount calculating unit 140, activity data storage unit 150, abnormal situation determination unit 160, abnormal situation notification unit 170 is included.

The first to nth photographing units 111, 112, and 113 are installed in a plurality of zones in an environment (ie, a house) where the elderly living alone are installed, and transmits images captured in real time to the image processor 120 in real time. do. At this time, at least one of the first to n-th imaging units 111, 112, and 113 according to an embodiment of the present invention is installed in an entrance area (eg, a porch) of the house. For reference, the image photographed through the photographing unit installed in the entrance area is used as a reference data for identifying the number of persons afterwards.

The image processor 120 receives images from the plurality of photographing units 11, 112, and 113 in real time, and divides each image into data in units of frames. In addition, the image processor 120 sequentially transmits the processed frames to the occupancy calculation unit 130 and the activity amount calculating unit 140. For reference, the image processor 120 transmits the images received from all the photographing units 111, 112, and 113 to the activity amount calculating unit 140, and is received from a specific photographing unit (that is, the photographing unit installed in the entrance area). The video may be transmitted to the occupancy calculation unit 130.

The human resources calculation unit 130 performs a predetermined image processing on the received image to determine whether a person is in and out, and calculates the number of human resources in the house based on the determination.

In detail, the occupant calculation unit 130 receives an image photographed through a photographing unit installed in an entrance area, and performs a predetermined image processing on the received image to identify whether a person in the corresponding area enters or exits. For reference, the occupant counting unit 130 may determine whether or not a person enters or exits by using various image processing methods such as object movement identification and facial recognition.

And, the number of people occupancy calculation unit 130 calculates the number of people in the current house based on whether or not the identified person is in and out, and if it is determined that the number of people in real time calculated or the number of people in the calculated number is predetermined number of abnormalities The situation determination unit 160 transmits the information of the current number of people in residence.

On the other hand, the number of people occupancy calculation unit 130 may transmit and update the number of people in the activity data storage unit 150, each time the number of people in the area changes. For reference, the activity data storage unit 150 to be described below may classify and store the number of in-service personnel transmitted from the in-service personnel calculation unit 130 by time zone. That is, the activity data storage unit 150 may update and store the changed number of people in one time range to the most recently changed number of people, but may store the number of people in time.

The activity calculation unit 140 calculates an activity amount of the object in the image based on the change amount of the pixel value in the frame of the received image, and transmits and stores the calculated activity amount to the activity data storage unit 150.

In detail, the activity calculation unit 140 calculates a difference of each pixel value between front and rear frames with respect to each image, and detects the number of pixels having a deviation greater than or equal to a predetermined reference deviation. For reference, the activity amount calculating unit 140 may calculate a change amount of the pixel value between frames based on at least one of attribute values such as color and brightness of the pixel.

The activity calculator 140 calculates the activity of the object by summing the number of the detected pixels within a preset time range. For example, when the predetermined time range is one hour, the activity amount is calculated by summing the number of the detected pixels for each time zone.

In this case, the activity amount calculating unit 140 may store, in the activity data storage unit 150, a value obtained by adding activity amounts calculated for each of a plurality of received images (that is, images of different zones) for each time period.

The activity data storage unit 150 classifies the activity amount calculated from the activity amount calculation unit 140 by date and time zone and stores the activity amount data. At this time, the activity amount data storage unit 150 may store the activity amount data further including the number of people in the calamity calculated from the occupancy number calculating unit 130.

For example, FIG. 2 is a diagram for explaining an example of an activity data storage method according to an embodiment of the present invention.

As illustrated in FIG. 2, the activity data storage unit 150 may store activity data including the values of the amount of activity and the number of people for a predetermined period, by time and date.

The activity data storage unit 150 detects and provides activity data requested from the abnormal situation determination unit 160. At this time, the activity data storage unit 150 is requested from a data structure that stores the activity data for a predetermined period of time (represented as 'one month' in Figure 2) as shown in Figure 2 in a circular queue (circular queue) method Activity data can be detected. As such, by storing the activity data in a circular queue manner, the storage space and data throughput can be efficiently minimized.

1 again, the abnormal situation determination unit 160 determines the occurrence of the abnormal situation in real time based on the activity amount calculated from the activity amount calculating unit 140 and the activity amount data previously stored through the activity data storage unit 150. do.

Specifically, the abnormal situation determination unit 160 compares the value of the same amount of activity on the other day and the calculated value of the activity amount among the previously stored activity amount data, and determines the occurrence of the abnormal situation based on the difference value between the compared activity amounts. To judge. In addition, the abnormality determination unit 160 compares the currently calculated activity amount with the activity amount calculated in the same day time zone, or compares the currently calculated activity amount with the average value of the same time activity amount within a predetermined reference period (for example, one month). Can be compared.

In this case, the abnormal situation determination unit 160 may determine that an abnormal situation occurs when the difference value between the compared activity amounts exceeds a predetermined deviation value. For example, when the amount of activity currently calculated is less than the predetermined deviation value compared to the amount of activity calculated during the same day, it means that the movement amount of the elderly living alone in the house is significantly reduced, so that an abnormal situation may occur.

On the other hand, the abnormal situation determination unit 160 may determine the occurrence of the abnormal situation on the basis of the number of people in the calamity calculated by the activity personnel calculating unit 130 together with the activity amount currently calculated from the activity amount calculating unit 140.

Specifically, the abnormal situation determination unit 160 obtains the number of in-country people calculated in real time, and determines the abnormal situation by comparing the calculated amount of activity with pre-stored amount of activity when the currently calculated number of in-country people is determined to be a preset number of people. do. That is, when the elderly living alone are at home alone, it is possible to detect a case where the amount of change in the amount of activity generated is more noticeable than other days.

If it is determined that the abnormal situation has occurred, the abnormal situation notification unit 170 outputs notification information informing that the abnormal situation has occurred.

Specifically, the abnormal situation notification unit 170 outputs notification information of a predetermined form (such as text or sound data) through a notification member (not shown) included in or separately connected to the abnormal situation detecting apparatus 100, or in advance. Wired / wireless communication data can be transmitted to management terminals such as government offices and hospital emergency rooms.

Hereinafter, an abnormal situation detection method according to an embodiment of the present invention will be described in detail with reference to FIG. 3.

3 is a flowchart illustrating an abnormal situation detection method according to an embodiment of the present invention.

First, an image of each zone is received in real time from a photographing apparatus (for example, a surveillance camera) installed in a plurality of zones (different zones) of a house where the elderly living alone live (S310).

In this case, at least one of the plurality of photographing apparatuses may be installed in the entrance area of the house.

Thereafter, the pixel change amount between frames is calculated for each acquired image (S320).

In this case, the difference of the value of each pixel of the front and rear frame of each image may be calculated, and the difference of the pixel value may be calculated based on at least one of attribute values such as color and brightness of the pixel.

Then, the amount of activity of the object in the image is calculated based on the amount of pixel change calculated per frame (S330).

Specifically, the number of pixels having a deviation value greater than or equal to a predetermined reference deviation is detected for each frame, and the amount of activity is calculated by summing the number of the detected pixels for each frame within a predetermined time range.

For reference, a value obtained by adding activity amounts calculated for a plurality of images (that is, images of different zones) for each of the same time periods may be calculated as the activity amount.

The calculated amount of activity is classified and stored by date and time zone.

Thereafter, an occurrence of an abnormal situation is determined by comparing the activity amount within a predetermined reference period with the currently calculated activity amount among the activity amounts for each time period of another date previously stored (S340).

Specifically, it may be determined that an abnormal situation occurs when a difference between the value of the same amount of activity on the other day and the value of the calculated amount of activity exceeds a predetermined deviation among previously stored amount of activity data.

Then, when the occurrence of the abnormal situation is determined, the notification information for notifying the occurrence of the abnormal situation is output (S350).

Meanwhile, in FIG. 3, in the abnormal situation detection method according to an embodiment of the present invention, the occurrence of the abnormal situation is determined based on the amount of activity, but is abnormal based on the number of people in the house where the elderly living alone together with the amount of activity. It is also possible to judge the occurrence of the situation.

That is, prior to performing the step (S340), the step of identifying the access of the person based on the image received from the photographing equipment installed in the entrance area in the house and calculating the number of people in the house according to the identified result, and The method may further include determining a case where the calculated number of occupants is a predetermined number (eg, one). As such, the steps S340 and S350 may be performed only when the calculated number of in-service personnel is a predetermined number of people.

The foregoing description of the present invention is intended for illustration, and it will be understood by those skilled in the art that the present invention may be easily modified in other specific forms without changing the technical spirit or essential features of the present invention. will be. It is therefore to be understood that the above-described embodiments are illustrative in all aspects and not restrictive. For example, each component described as a single entity may be distributed and implemented, and components described as being distributed may also be implemented in a combined form.

The scope of the present invention is shown by the following claims rather than the above description, and all changes or modifications derived from the meaning and scope of the claims and their equivalents should be construed as being included in the scope of the present invention. do.

100: abnormal situation detection device
111,112,113: Shooting Department
120:
130: resident personnel calculator
140: activity calculation unit
150: activity data storage
160: abnormal situation determination unit
170: Abnormal situation notice part

Claims (8)

In the abnormal situation detection device,
A plurality of photographing units installed in a plurality of zones in the house and respectively photographing images;
Computing the difference of each pixel value between the front and rear frame of the captured image to detect the number of pixels having a deviation or more than a predetermined reference deviation, and to calculate the amount of activity of the object by adding the number of the detected pixels within a predetermined time range Activity calculation unit;
An activity data storage unit classifying the calculated activity amount by date and time zone and storing the calculated activity amount as activity amount data; And
An abnormal situation determination unit for determining the occurrence of an abnormal situation in real time based on the calculated activity amount and the activity data stored in advance,
The abnormal situation determination unit,
And comparing the value of the activity amount of the same time zone with the calculated activity amount among the previously stored activity amount data and determining the occurrence of the abnormal situation based on the difference value between the compared activity amounts. The method of claim 1,
The abnormal situation determination unit,
And comparing the calculated amount of activity with the amount of activity calculated in the same time zone the previous day, or comparing the calculated amount of activity with an average value of the amount of activity in the same time period within a predetermined period. The method of claim 1,
The apparatus further includes a weighted person calculation unit configured to determine whether a person is in or out by performing a predetermined image process on the image received from the photographing unit installed in the entrance area of the house, and calculate the number of people in the house based on the determination. ,
The abnormal situation determination unit,
An abnormal situation detection apparatus for performing the determination process of the occurrence of the abnormal situation under the condition that there is one person in the house. The method of claim 1,
If it is determined that the abnormal situation has occurred, and further comprises an abnormal situation notification unit for outputting notification information informing of the occurrence of the abnormal situation, abnormal situation detection apparatus. In the abnormal situation detection method through the abnormal situation detection device,
(a) acquiring images captured in real time from a photographing apparatus installed in a plurality of areas within a house;
calculating a change value of each pixel value between front and rear frames of the obtained image;
(c) detecting the number of pixels whose calculated change value is equal to or greater than a predetermined deviation value;
calculating an amount of activity of an object by accumulating the number of pixels detected within a preset time range;
(e) storing the calculated amount of activity to correspond to a previously classified date and time zone;
(f) comparing the value of the previously stored activity amount with the value of the calculated activity amount; And
(g) determining an occurrence of an abnormal situation based on the difference value between the compared amounts of activity,
The step (f)
And detecting the comparison between the calculated activity amount and the activity amount in the same time period among the previously stored activity amounts, and detecting the abnormal situation. The method of claim 5, wherein
The step (f)
Detects the amount of activity of the same time within the previous day or a predetermined period of the amount of activity stored in advance,
And detecting a mean value of the detected amount of activity and a value of the calculated amount of activity when detecting the amount of activity in the same time period within the predetermined period. The method of claim 5, wherein
Before the step (b)
Determining whether a person is in or out by performing a predetermined image process on the image received from the photographing unit installed in the entrance area of the house, and
Comprising the step of calculating the number of people in the house based on the determination,
And performing the steps (c) to (g) when the number of occupants is one as a result of the calculation. The method of claim 5, wherein
After step (g),
If it is determined that the abnormal situation has occurred, further comprising the step of outputting the notification information informing of the occurrence of the abnormal situation, abnormal situation detection method.

Images