KR101514061B1 - Wireless camera device for managing old and weak people and the management system thereby - Google Patents

Abstract

The present invention relates to a camera capable of automatically determining and responding to various emergency situations including a fire occurrence for the protection and management of the elderly living alone, and an elderly management system using the same.
To this end, the present invention provides a portable camera device having a built-in battery unit, a target object detection unit for detecting a movement of a target object and a target object from the captured image, A fire detection unit for detecting a fire and an emergency call unit for using a button when the elderly person asks for help.
The target object detection unit detects an object of interest by an Adaboost method and extracts vision information of the object using the Kalman filter to identify the target object and recognize a situation according to the motion. The fire detection unit extracts and separates the moving object from the photographed image using the Gaussian blending model, and detects the target object using the HSL color saturation and lightness based on the HSL color space to determine whether or not the fire is present .
Also, an ultra-wide-angle mirror unit having a hyperboloidal reflection surface, which is located opposite to a photographing unit of the camera apparatus and has a horizontal section in the form of a hyperbolic function, is provided to acquire a photographed image having a wider angle of view.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a wireless camera device for managing elderly people,

The present invention relates to a wireless camera for protecting and managing the elderly living alone and an elderly management system using the same. More specifically, the present invention can manage the elderly's status in real time regardless of the location of the elderly living alone, The present invention relates to a camera capable of automatically determining and coping with a situation and an elderly management system using the same.

Unlike in the past, where the elderly living with their family members live together in an emergency, they need a system to observe and manage the health of the elderly living alone. .

However, since the conventional CCTV is fixedly installed, it is difficult to change the position of the CCTV. In order to observe a wide range due to the limited angle of view of the CCTV itself, a large number of cameras have to be installed.

In addition, since a separate observer must directly observe the screen periodically, if an emergency situation does not occur immediately, it may not be possible to take action within a reasonable period of time. Furthermore, since the conventional CCTV is used only for observation through a camera, communication between an observer and a visitor is difficult.

Also, in the case of the conventional CCTV, the angle of view is generally about 60 degrees, and there is a problem that the state of the elderly can not be grasped when the elderly person is located outside the angle of view.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a management camera that is not disposed in a fixed form but can be easily moved and installed, and capable of bidirectional communication between an observer and a target observer.

Furthermore, the present invention aims to automatically determine an emergency such as a fire, and to specify and observe a passenger in a photographed image so as to promptly take measures.

It is yet another object of the present invention to provide a camera capable of observing the surrounding environment without a square area in a screen in which the photographing angle of view is wider than that of a typical CCTV camera.

According to an aspect of the present invention, there is provided an elderly care camera device including a battery unit, the camera device comprising: a photographing unit that photographs a surrounding image of the camera device and converts the peripheral image into electrical image information; A wireless communication unit for transmitting and receiving information to and from a wireless intermediate apparatus located outside the camera apparatus; A microphone unit for recording the surrounding voice as electric voice information; A speaker unit for transmitting sound to the surroundings; An emergency call unit provided outside the camera apparatus with a button shape for transmitting an emergency signal to the wireless mediator via the wireless communication unit; A recording unit for recording the image information and the audio information as moving pictures in a detachable video recording unit; A target object detection unit for detecting a motion of a target object and a target object from the photographed image; And a fire detection unit for detecting a fire in a peripheral image of the camera device from the image information.

Wherein the target object detection unit comprises: a video decoder unit for converting the photographed image into digital information; And an image extracting unit for extracting motion detection and motion regions from the information obtained from the image decoder unit, detecting an object of interest by the Adaboost method in the region, and extracting the vision information of the object using the Kalman filter And an information processing unit (DSP). Here, the vision information is face information or behavior information of an object.

Wherein the camera apparatus further comprises a fire detection unit for detecting a fire in a peripheral image of the camera apparatus from the image information, wherein the fire detection unit detects at least one A storage unit for storing at least one frame sequence including a fixed object corresponding to a background image, extracting and separating a moving object of the moving object from the photographed image photographed by the photographing unit; And a sensing unit based on a Hue Saturation Light Color Space and sensing the object using saturation and lightness of the HSL. The fire detection unit further includes a setting unit for setting a reference value for separating the moving object based on a probability density of a specific pixel along a Gaussian distribution.

The present invention also relates to a plurality of elderly care camera devices; A wireless mediator connected to the plurality of elderly care camera devices via a wired / wireless network to enable information communication; And an elderly management server, wherein the elderly management server comprises: a wired / wireless transmission / reception unit for receiving image information and voice information from the plurality of elderly care camera devices; A recording unit for recording the transferred image information and audio information as moving pictures; A target object detection unit for detecting a motion of a target object and a target object from the photographed image; And a fire detection unit for detecting a fire in a peripheral image of the camera apparatus from the image information.

According to the present invention, since the camera device can be movably arranged, it is possible to easily install the camera device at a desired position to manage the elderly person living alone, and to transmit and receive voice remotely, thereby enabling bidirectional communication between the observer and the visiting user.

In addition, it is possible to automatically judge the condition of the person to be concealed through the movement of the person to be abruptly collapsed, etc., and can quickly take measures against the emergency by automatically judging whether or not the fire has occurred.

In addition, when the elderly person performs strange behavior in the management of the elderly person by putting the recording function, the recording function can be activated by pressing the record button on the family's smartphone outside, and it is possible to display the usual recorded action Your doctor may help you with your prescription.

In addition, if the elderly person needs help by inserting the emergency call function, the emergency call function can be activated by pressing the emergency call button, and this can be transmitted to the elderly management server to inform that it is an emergency.

1 shows a camera device for managing elderly people according to the present invention.
2 is a flowchart illustrating a method of detecting a motion of a target object and a target object in a target object detection unit.
3 shows an elderly care system according to the present invention.
4 shows components of the fire detection unit.
5 is a view illustrating an embodiment of extracting a background image using a Gaussian mixture model in a fire detection unit.
6 is a view showing an embodiment for detecting smoke based on the HSL color space in the fire detection unit.
7 is a diagram illustrating a process in which data is transmitted and received between the components included in the fire detection unit of FIG.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present invention. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Also, in order to clearly illustrate the present invention in the drawings, portions not related to the present invention are omitted, and the same or similar reference numerals denote the same or similar components.

The objects and effects of the present invention can be understood or clarified naturally by the following description, and the purpose and effect of the present invention are not limited by the following description.

The objects, features and advantages of the present invention will become more apparent from the following detailed description. In the following description, well-known functions or constructions are not described in detail since they would obscure the invention in unnecessary detail.

1, an elderly care camera device 100 according to the present invention will be described.

The portable camera device for managing the elderly according to the present invention includes a photographing part 110 photographing a surrounding image of the camera device and converting it into electrical image information; A wireless communication unit 130 for transmitting / receiving information to / from a wireless intermediate apparatus located outside the camera apparatus; A microphone 140 for converting the surrounding voice into an electrical voice signal; And a speaker unit 150 for transmitting sound to the surroundings.

The photographing unit 110 is, for example, a lens and an image sensor for converting a surrounding environment into electrical image information. According to a preferred embodiment of the present invention, the lens of the camera has an angle of view of about 60 degrees or more, (Preferably about 190 degrees) using an ultra-wide angle lens including a fisheye lens, and the image sensor can be a CCD or a CMOS image sensor of a camera that acquires a photographed image as an electrical signal have. The photographing unit 110 converts an image of an object located within a front view angle into an electrical image form, and becomes an object of various information processes to be described below.

The wireless communication unit 130 is built in the camera device according to the present invention and receives information from outside or transmits information to the outside. According to a preferred embodiment, image information (video information) or voice information is transmitted to an external wireless intermediate apparatus 200 (for example, an access point (AP) for a Wi-Fi, a router, (Image information) from an external device (an external terminal via the Internet or the Internet) through the Internet 200.

The microphone unit 140 is built in the camera device and transmits sound or voice generated in the surrounding environment as acoustic information to the outside through the wireless communication unit. The speaker unit 150 outputs sound information received from the outside, It is used for alarm output according to the situation.

Preferably, the camera apparatus 100 includes a battery unit (not shown) so that the camera apparatus 100 can be driven even if power is not supplied from the outside. Therefore, since the battery is configured with the wireless communication unit 130, the transmission and reception of the telegram and the power supply are performed without any connecting means from the outside, so that the portability can be maximized.

The recording unit 180 receives a recording start signal transmitted from the external terminal (the elderly management server 300 or the management terminal 400) of the camera apparatus 100 to be described later through the wireless communication unit 130, The moving picture storage unit 185 stores the video information transmitted from the photographing unit 110 and the sound information transmitted from the microphone unit 140. The moving picture storage unit 185 preferably removes As a possible flash memory, for example, an SD card, a CF card, a memory stick, or the like, but includes all means capable of storing moving pictures. When the elderly person exhibits abnormal behavior through such a video recording function, The terminal 400 transmits a recording start signal to record an abnormal behavior and then displays a recorded abnormal behavior when a diagnosis is received by a doctor, It can be helpful.

The emergency call part 120 is preferably provided outside the camera device 100 in a button manner so that the emergency call part 120 is pressed by the wireless communication part 140 And transmits the emergency signal to the elderly management server 300 or the external management terminal 400 through the wireless mediation apparatus 200 through the wireless mediation apparatus 200. Thus, an urgent signal is immediately transmitted to the elderly management server 300 or the external family terminal 400, so that it is possible to visually check whether an actual emergency situation has occurred through the photographing unit 110, It is possible to respond promptly in an emergency situation.

When the emergency signal is transmitted to the external terminal through the emergency call unit 120, the recording unit 180 automatically transmits the emergency signal through the photographing unit 110 and the microphone unit 140, The surrounding video may be left as a moving picture in the moving picture storage unit 185 so that the old person may leave his / her own symptom as a moving picture before the consciousness is lost, , So that the doctor can refer to the video and take more appropriate measures.

How to determine movement and status of target object

According to a preferred embodiment of the present invention, the target object detecting unit 170 may further include a target object detecting unit 170 for identifying a target object (for example, the elderly living alone) or determining and tracking the movement thereof. The target object detection unit 170 includes an image decoder unit 171 for converting the photographed image image into digital information for detecting the motion of the target object and the target object from the photographed image, And an image information processing unit 173 for extracting image information. More specifically, the image information processing unit 173 extracts a motion detection and motion region from the information obtained from the image decoder unit 171, detects an object of interest using the Adaboost method in the region, And extracts the vision information of the object using the Kalman filter.

Hereinafter, a method of detecting a motion of a target object and a target object in the target object detection unit will be described with reference to FIG.

The flow chart of FIG. 2 is a flowchart illustrating the steps of obtaining image information (S110), motion detection and motion region extraction step S120, object detection step S130, motion estimation and tracking step S140, And a notification step (S160).

The motion detection step (S120) comprises background scene registration, motion detection, and motion region separation. Registration of the background image registers the average value of the background image and the standard deviation of the degree of change by learning the background image. We used the moving average model to adapt the change of the registered background image according to the diary change of the day. The motion detection is performed in the pixels of each frame. When the difference between the pixel value of the current screen and the pixel value of the registered background screen is equal to or greater than a predetermined multiple of the registered standard deviation value, it is determined that the pixel has motion. For example, if the standard deviation value is more than a certain multiple of the standard deviation value, it can be determined that an object is moving in a space (for example, a room) where the camera device according to the present invention is located. According to a preferred embodiment, after detecting motion in each pixel, a morpholgy filter may be applied to reduce the noise effect. When motion detection is performed, a histogram representing the number of motion pixels in the horizontal and vertical directions of the image is generated, and a separate motion object is separated by applying a histogram to the Gaussian mixture model.

In the object detection step (S130), a "Haar-based object detection method ", which is known as one of the best object detection methods, is applied. This method is a very effective method of detecting the object by learning the object to be detected by the Adaboost method. Since the calculation amount is large in order to detect the object in real time in the moving picture, the object is detected only in the motion region extracted from the previous step (S120) Respectively. The surveillance system according to the present invention to be described below can be configured to monitor a wide area, and since the corresponding area occupies only a part of the entire image, the present invention is very effective.

In the motion estimation and tracking (S140), a known 3D Kalman filter image motion estimation method is applied. The main purpose of the present invention is to correctly estimate motion in each pixel in order to determine whether the motion estimation itself is not an object but the next step (S150).

The situation recognition step (S150) depends on the application area of the camera device according to the present invention. If the elderly living alone show movement in a certain space (for example, a room) and then stop moving for a certain period of time, it can be recognized that the elderly person has fallen into a state of unconsciousness. In other words, when a person falls, he / she may be able to move for a very short period of time, and since the movement may then be stopped continuously, if the two conditions are met, This can reduce misjudgment that can occur.

If it is determined that the specific situation has already been set after the state recognition step, the emergency state notification step (S160) may be performed by an external terminal (for example, a management terminal to be described later).

Senior management server and system

According to a preferred embodiment of the present invention, there is provided an elderly management system (10) which has a plurality of camera devices according to the present invention and manages one elderly person on various screens or manages a large number of elderly people in a town where the elderly living alone lives (See FIG. 3).

The elderly management system 10 includes a plurality of camera devices 100 according to the present invention, a plurality of target objects (for example, an elderly person) receiving image information or situation recognition information from the plurality of camera devices 100, The elderly management server 300 that tracks the current position and state of the elderly person by tracking and identifying the elderly person, and a management terminal 400 that can confirm elderly management results. Here, the elderly management server 300 and the management terminal 400 may exist in an integrated form.

In order to specify the current position of the plurality of camera apparatuses 100, the respective camera apparatuses may include the positioning means 160. Although the positioning means 160 is preferably GPS, it is not limited to this, and it is possible to use positioning means for measuring the current position from the position of one or more networks to which the present camera apparatus is connected.

The elderly management server 300 according to the present invention may include a wired / wireless transmission / reception unit that receives image information or context information from a plurality of elderly management camera apparatuses 100. The wired / wireless transmission / reception unit may receive various information transmitted from the elderly management camera apparatus 100 through the wireless mediator 200, or may transmit various control information to the elderly care camera apparatus 100.

The elderly management server 300 may preferably include a target object detection unit 370 used in the elderly care camera apparatus 100. Therefore, without having to include the target object detection unit 170 in each elderly care management camera apparatus 100, the camera apparatus simply transmits the photographed image information to the management server, and detects the movement of the target object and the target object in the transmitted image information And a process of detecting the situation and recognizing the situation can be integrally performed in the server. According to the present embodiment, it is not necessary to provide the target object detection unit 170 in each camera apparatus, so it is possible to reduce the size of the camera apparatus and the manufacturing cost, and the image processing is performed integrally in the server, . However, if the management server fails, the entire system may become inoperable. Therefore, each of the camera apparatuses 100 may be provided with the target object detection unit 170 so as to detect the situation even in an environment in which the server does not operate or is not connected to the server .

Also, the elderly care management server 300 may include a recording unit 380 used in the elderly management camera apparatus 100. Accordingly, even though the moving image storage unit is not attached to the elderly management camera apparatus 100, the server 300 can store the transmitted image information and the audio information, thereby recording the movement of the target object and the target object. Since the server 300 generally includes a mass storage unit (not shown), it is not necessary to separately provide a detachable video storage unit unlike the recording unit 180 in the camera apparatus.

In the elderly management system, it is possible to specify the target object through the target object detection unit 170 (370). Even if the target moves away from the place where the user himself / herself resides, 100) located in the region where the image is captured, the movement path of the target object can be grasped. Therefore, even if the demented elderly person moves out of his / her residence and moves to another place, management of the demented elderly can be facilitated because the present location of the demented elderly person can be specified if the plural camera devices are located in the shooting region. For example, if the position of the target object is outside the position range information based on the predetermined range information of the target object (demented elderly), the state recognition step of the target object detection unit (S150) .

According to a preferred embodiment, the elderly management system according to the present invention includes at least one management terminal 400 capable of confirming the elderly management result, and the management terminal 400 judges And is connected to the management server 300 through a network such as the Internet to receive the management result. The management result is notified in the form of an alarm, and if the above-mentioned emergency occurs, it is transmitted through telephone connection or text notification.

According to a preferred embodiment, the management terminal 400 can be connected to the management camera device directly or via the elderly management server 300 as a smart phone. In this case, bidirectional communication may be possible through the speaker unit and the microphone unit provided in the camera apparatus. Therefore, as soon as it is notified that it is an emergency, the current status of the target elderly person can be confirmed by telephone connection.

Also, as described above, when the elderly man performs an unusual behavior in the elderly management, he or she can give a recording instruction to the recording unit of the elderly care camera device 100 from the outside of the family management terminal 400, By showing the usual recorded actions when receiving a prescription, the doctor may be helpful in making the prescription.

If it is determined that the elderly person needs help through the emergency call unit 120, the emergency call function can be activated by pressing the emergency call button. The elderly management server 300 or the family management terminal 400 ) To inform that it is an emergency.

Fire detection unit

According to a preferred embodiment, the camera apparatus according to the present invention may further include a fire detection unit 190 for detecting a fire in a peripheral image of the camera apparatus from the image information. According to another embodiment, the fire detection unit 190 may additionally or alternatively be included in the elderly management server 300. That is, although it is preferable that the fire detection unit 170 is included in each camera device 100 like the target object detection unit 170, the camera sends only the image image to the server, The convenience can be enhanced.

Hereinafter, the fire detection unit 190 will be described in more detail.

The fire detection unit 190 can extract a background image including a fixed object using a Gaussian Mixture Model. In this case, a fixed object is defined as a background image. For example, assuming that the background is a room, except for a moving object such as a moving object in a room and a shadow for the object, Quot; image " At this time, the moving object may be a moving object, or a shadow of an object.

The fire detection unit for fire detection uses HSL saturation (S: Saturation) and lightness (L: Light) based on the HSL color space (Hue, Saturation, Light Color Space) . That is, among the image images photographed by the photographing unit 110, in order to distinguish only the fire smoke from the similar object (smoke generated when food is naturally generated), the fire detection unit for detecting fire includes HSL color space analysis .

FIG. 4 is a block diagram illustrating components of the fire detection unit 190 for fire detection. FIG. 5 is a flowchart illustrating a method of extracting a background image using a Gaussian mixture model in a fire detection unit 190 (FIG. FIG. 6 is a view showing an embodiment of detecting smoke based on the HSL color space in the fire detection unit 190. FIG. Although the embodiment shown in this drawing is described on the assumption that a forest fire has occurred, the method according to the present invention can also be applied to a method of detecting a fire occurring in a house.

Referring to FIG. 4, the fire management unit 190 for detecting fire includes a storage unit 191 and a sensing unit 193. Preferably, the setting unit 195 may further include a setting unit 195. Here, the storage unit 191 extracts and separates at least one moving object except for the object corresponding to forest fire from the photographed image photographed by the photographing unit 110, using a Gaussian Mixture Model , And stores at least one frame sequence corresponding to the background image and including a fixed object. At this time, the destination object may be a smoke generated in a fire.

Hereinafter, a method of storing a background using a Gaussian mixture model will be described.

The Gaussian blend model is one of the reliable methods for extracting objects and storing background images for a certain period of time. In this case, the Gaussian mixture model can be used to store a background image to automatically detect an object such as smoke, and a background subtraction method can be used to extract an object in a frame sequence. Also, the image pixels in the frame sequence follow a Gaussian distribution, in which the corresponding pixel values can be varied in various ways corresponding to the frame sequence.

The pixel state in the frame sequence is defined as k, and k is defined as k {1,2, ..., n}. k can have a value from 3 to 7, which is referred to as a multi-modal distribution. The pixel value X in the pixel processing can be changed based on the frame sequence state of the image. At this time, it is assumed that the pixel processing has a normal distribution of the following equation (1): f (Probability Density Function).

Where X is a random variable of the pixel change, [mu] _k is the average of the pixel state k, and [Sigma] _k is the variance of the pixel state k.

Referring to FIG. 5A, Variable Distribution of a specific pixel in a frame sequence according to time is shown. In this case, k is a pixel state variable in the frame sequence. For example, the pixel state probabilities to be changed in 100 frames are p (k = 1), p (k = 2), and p (k = 3). The pixel probability density is p (k = 2), (c) the pixel at the steady state passes through the dark shadow, The probability density is p (k = 3), where the sum of the probabilities is 1, ie p (k = 1) + p (k = 2) + p (k = 3) = 1.

At this time, when the pixel processing variation is stable, that is, in a stationary process, a maximum likelihood function can be obtained by an Expectation Maximum (EM) algorithm. At this time, the parameter can be estimated by a K-mean (K-mean) approximation method.

The setting unit 195 can set a reference value for separating the moving object based on the probability density of a specific pixel following the Gaussian distribution.

The k value can determine the current state, and if a new image frame is input, the initial state can be updated based on the variance corresponding to the background pixel, which can be calculated by the following equation (2).

는 시간 t에서의 k 상태 확률의 합이고, α는 학습률이다. 또한, M_k,t+1은 0 또는 1의 값을 가지는데, 만약 ωk,t+1 이 ω_k,t 일치하는 경우 M_k,t+1은 1이고, 일치하지 않는 경우 M_k,t+1은 0이다.Where ω _{k, t} = p (k) is the k-state probability at frame t,

Is the sum of the k-state probabilities at time t, and a is the learning rate. In addition, M _{k, t + 1} is takes a value of 0 or 1, if ωk, t + 1 is ω _k, if they match _t M _{k, t + 1} is 1, if they do not match M _{k, t +1} is zero.

Also, when a new frame is processed, the mean [mu] and variance [Sigma] are also recalculated by the following equations (3) and (4). At this time, if ω _{k, t + 1} coincides with ω _{k, t} , the current state probability of new frame (t + 1) is the same as the previous state of time t. Conversely, if they do not match, the current state of the new frame may be updated by a learning rate?.

Referring to FIG. 5B, an image processing for estimating a background based on a Gaussian mixture model is shown. In FIG. 5 (a), the right side is dark in the center of the figure due to the shadow of the cloud. In the case of extracting the background based on the Gaussian mixture model, only the background in which the shadow is blurred as shown in (b) and the shadow is excluded as shown in (c) finally can be extracted.

Where B _{k (0? K? N)} is the background image sequence stored in the buffer, I is the pixel intensity, and (X, Y) is the image size. At this time, the fire detection unit according to an embodiment of the present invention may be configured to store background images at intervals of 60 seconds.

In this equation, B1 is the background image stored in the buffer in 60 seconds, and B2 is the background image stored in the buffer in 120 seconds. The stored background image frame sequence may be expressed as {B ₀ , B ₁ , B ₂ , ..., B _N }.

Referring to FIG. 5C, a frame sequence of a normal image including a fixed object, that is, a background image in which a moving object is removed in a situation before a fire, a background image in an abnormal state in which smoke is generated, (Subtraction Image).

Referring back to FIG. 4, the sensing unit 193 senses the target object based on the Hue Saturation Light Color Space and using saturation and lightness in the HSL. That is, among the images photographed by the photographing unit 110, the HSL color space analysis is used to distinguish only the fire smoke.

At this time, the sensing unit 193 resets the saturation and the brightness range from 0 to 255 degrees, and based on the adjusted saturation and brightness, the saturation of the background image is 0 to 60 degrees, and the brightness of the background image is 40 to & 140 degrees, it can be detected that the object object exists in the background image. At this time, the destination object may be an initial state of fire.

On the other hand, the color space can be defined as means for proceeding with the specification, generation and visualization of colors. At this time, color image processing using a computer can be based on various combinations of RGB (Red, Green Blue) phosphors required for forming colors. However, since the false alarm can be generated when the RGB color space processing technique is used, the fire detection unit 190 for fire detection according to an embodiment of the present invention uses the saturation and brightness to detect smoke.

At this time, the HSL has three values. That is, first, the hue element has a range of 0 to 360 degrees and can implement the color itself in the form of a degree. In this case, 0 indicates red, 120 indicates green, 240 indicates blue, 60 indicates yellow, and 300 indicates magenta.

Second, the Saturation element has a percentage value. For example, 100% is defined in full color. Light elements, like chroma, have a percentage value, for example 0% means dark or black, 100% is light, that is white, and 50% It means middle.

In the case of the fire detection unit 190 for fire detection according to an embodiment of the present invention, the saturation and brightness having a range of 0 to 100% are divided into a range of 0 to 255 degrees Can be readjusted. Referring to FIG. 6, the saturation and brightness of the smoke (a) are distributed in the range of 0 to 60 and 40 to 140, respectively, and the saturation of the fog (b) ranges from 0 to 80, , But it can be seen that the brightness of the fog has a range of 140 to 220 different from smoke. (c) the saturation of the phosphorus clouds is 120 to 250 degrees and the brightness is 180 to 250 degrees.

If a fire is detected in the fire detection unit 190, that is, when a target object is detected, the wireless communication unit 130 according to the present invention transmits the fire alarm to the server 300 or the terminal 400 via the wireless mediation apparatus 200, (For example, an elderly management server or an elderly management terminal). According to a preferred embodiment of the present invention, when a fire is detected by the fire detection unit 190, the fire detection unit 190 may be configured to notify the user of the fire through the speaker unit 140.

Accordingly, the fire detection unit 190 according to an exemplary embodiment of the present invention primarily extracts a background using a Gaussian blending algorithm, and secondly, based on the HSL color space technique, It is possible to detect the smoke caused by the fire quickly and accurately. Accordingly, it is possible to accurately and quickly detect smoke as a source of the initial fire, thereby generating a warning message for the fire, thereby providing a basis for suppressing the fire in the early stage.

FIG. 7 is a diagram illustrating a process in which data is transmitted and received between the components included in the fire detection unit 190 of FIG. 4 according to an embodiment of the present invention. Hereinafter, an example of a process of transmitting and receiving a signal according to an embodiment of the present invention will be described with reference to FIG. 7, but the present invention is not limited to such an embodiment, and according to various embodiments described above It is apparent to those skilled in the art that the process of transmitting and receiving data shown in FIG. 7 can be changed.

Referring to FIG. 7, the photographing unit 110 of the camera device 100 captures an image (S210), and transmits the captured image information to the fire sensing unit 190 (S220). Then, the fire detection unit 190 receives the photographed image in real time or at regular intervals (S230), and extracts the moving object excluding the target object from the photographed image using the Gaussian mixture model (S240).

In addition, the fire detection unit 190 for fire detection stores at least one frame sequence corresponding to the background image and including the fixed object in the storage unit 191 (S250), and the color saturation S The target object is sensed using the brightness L (S260).

In this case, if at least one of the at least one frame sequence has a saturation of 0 to 60 and a brightness of 40 to 140 (S270), the at least one pixel is detected as an object object and an alarm message is generated )do.

If the pixel having the saturation of 0 to 60 and the brightness of 40 to 140 is not present in at least one frame sequence in step S270, the process returns to the step of receiving the photographed image from the camera device 100. [

Finally, the fire alarm message is transmitted to the external device 300 (400) (S290). At this time, if the speaker unit 140 of the camera apparatus 100 sends an alarm message, the step S290 may be deleted.

The order between the above-described steps S210 to S290 is merely an example, but is not limited thereto. That is, the order between the above-described steps S210 to S290 may be mutually varied, and some of these steps may be executed simultaneously.

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 inventive concept as defined by the appended claims. But is not limited thereto.

In the above-described exemplary system, the methods are described on the basis of a flowchart as a series of steps or blocks, but the present invention is not limited to the order of the steps, and some steps may occur in different orders or simultaneously . It will also be understood by those skilled in the art that the steps shown in the flowchart are not exclusive and that other steps may be included or that one or more steps in the flowchart may be deleted without affecting the scope of the invention.

100: elderly care camera device 110: photographing part
120: emergency call section
130: wireless communication unit 140: microphone unit
150: speaker unit 170: target object detection unit
171: Video decoder unit 173: Video information processor
180: camera side recording unit 185: moving image storage unit
190: fire detection unit 191:
193: Detecting unit 195: Setting unit
200: Wireless Intermediary Device
300: Senior management server 370: Target object detection unit
380: recording section 390: fire detection section
400: terminal for management

Claims (7)

A camera device for an elderly care which is portable with a built-in battery part,
A photographing unit photographing a surrounding image of the camera device and converting the captured image into electrical image information;
A wireless communication unit for transmitting and receiving information to and from a wireless intermediate apparatus located outside the camera apparatus;
A microphone unit for recording the surrounding voice as electric voice information;
A speaker unit for transmitting sound to the surroundings;
An emergency call unit provided outside the camera apparatus with a button shape for transmitting an emergency signal to the wireless mediator via the wireless communication unit;
A recording unit for recording the image information and the audio information as moving pictures in a detachable video recording unit;
A target object detection unit for detecting a motion of a target object and a target object from the photographed image; And
And a fire detector for detecting a fire in a peripheral image of the camera device from the image information. A plurality of elderly care camera devices according to claim 1; A wireless mediator connected to the plurality of elderly care camera devices via a wired / wireless network to enable information communication; And an elderly management server,
The elderly management server includes:
A wired / wireless transmission / reception unit for receiving image information and voice information from the plurality of elderly care camera devices;
A recording unit for recording the transferred image information and audio information as moving pictures;
A target object detection unit for detecting a motion of a target object and a target object from the photographed image; And
And a fire detection unit for detecting a fire in a peripheral image of the camera apparatus from the image information. The method according to claim 1,
Further comprising position determining means for determining position information of the elderly care management camera device. The method according to claim 1,
Wherein the target object detection unit comprises:
An image decoder for converting the photographed image into digital information; And
An image decoder unit for extracting motion detection and motion regions from information obtained from the image decoder unit, detecting an object of interest using the Adaboost method in the region, and extracting vision information of the object using the Kalman filter And a processor (DSP). 5. The method of claim 4,
Wherein the vision information is face information or behavior information of the object. The method according to claim 1,
Wherein the fire detection unit comprises:
Extracting and separating at least one moving object except for the object corresponding to the fire from the photographed image photographed by the photographing unit using the Gaussian blending model and extracting at least one frame including the fixed object corresponding to the background image A storage unit for storing a sequence; And
And a sensing unit based on a Hue Saturation Light Color Space and sensing the target object using saturation and lightness of the HSL. . The method according to claim 6,
Wherein the fire detection unit comprises:
Further comprising a setting unit configured to set a reference value for separating the moving object based on a probability density of a specific pixel following the Gaussian distribution.

Images