KR20220075649A - Senior citizen access monitoring system - Google Patents

Abstract

The present invention relates to an access monitoring system for the elderly and more particularly, a camera unit for acquiring an image; a sensor unit installed on one side of the camera unit to obtain a user's access record; a storage server unit for storing the object image obtained from the camera unit and the user's access record; a communication unit interworking with the storage server unit for wired/wireless communication of object images and user's access records to the manager; A control unit for controlling the camera unit, the sensor unit, the storage server unit, and the communication unit; includes, wherein the control unit is controlled through an artificial intelligence algorithm.

Description

Senior citizen access monitoring system

The present invention relates to an access monitoring system for the elderly, and more particularly, to a system for monitoring the access of the elderly and infirm through a camera unit and a sensor unit for a guardian to check in real time whether the elderly are safe and whether to go out to perform an immediate action.

In recent years, the number of elderly living alone is increasing significantly as the rapid aging of the population and the advancement of living patterns are intertwined. Also, due to the increase in dual-income earners, the lifestyle in which only infants and babysitters live at home is also gradually increasing. In addition, the socially disadvantaged, such as the disabled, are vulnerable to living alone, so the need for management is urgent.

Means to more effectively protect the elderly, infants, and disabled people by resolving social neglect are being studied in a variety of ways. Therefore, it is practically difficult to check personal problems, so it is only at the level of checking in by visiting or using the phone for a certain period of time, so management efficiency is very poor.

As a conventional technology that enables continuous monitoring of the elderly, infants, and the disabled (hereinafter referred to as household members) living alone, there is Republic of Korea Patent Publication No. 10-2015-0116206 (hereinafter referred to as "prior art"). The technology discloses a method of notifying an emergency by measuring the vital signs of the elderly living alone.

Specifically, a wearable type biometric device is attached to the user, and biometric information such as heart rate, blood pressure, and body temperature is measured and transmitted to the user's own smartphone or emergency center. Here, the user's smart phone and the wearable type biometric device are connected through Bluetooth communication, and the smart phones and biometric devices of surrounding people are also configured to perform Bluetooth pairing.

However, even if there is no problem in notifying the emergency center using the user's smartphone, it takes a considerable amount of time for the emergency center to identify the user's location and move to the user, making it difficult to provide proper first aid or rescue. It is currently being discussed.

Republic of Korea Patent Publication No. 10-2015-0116206

The present invention has been devised in view of the above problems, and an object of the present invention is to check the safety of the elderly and the infirm through a camera unit and a sensor unit in real time, and a guardian access monitoring system for performing immediate actions is to provide

Another object of the present invention is to provide an access monitoring system for the elderly and infirm to perform an immediate action by notifying the guardian after quickly recognizing the entrant and at the same time being able to accurately recognize the elderly and infirm through the object recognition algorithm.

The purpose of the embodiments of the present invention is not limited to the above-mentioned purpose, and other objects not mentioned will be clearly understood by those of ordinary skill in the art to which the present invention belongs from the description below. .

According to a feature for achieving the above object, the present invention is a camera unit for acquiring an image;

a sensor unit installed on one side of the camera unit to obtain a user's access record;

a storage server unit for storing the object image obtained from the camera unit and the user's access record;

a communication unit interworking with the storage server unit for wired/wireless communication of object images and user's access records to the manager;

a control unit for controlling the camera unit, the sensor unit, the storage server unit, and the communication unit;

including,

The control unit may be provided with an access monitoring system for the elderly controlled through an artificial intelligence algorithm.

In addition, according to an embodiment of the present invention, the camera unit,

Recognizing a plurality of objects from the object image, the elderly access monitoring system further comprising an object recognition unit for obtaining the recognized object image may be provided.

In addition, according to an embodiment of the present invention, the object recognition unit,

An access monitoring system for the elderly and weak for performing object recognition using at least one of an SSD algorithm, a YOLO algorithm, and an AdaBoost algorithm may be provided.

In addition, according to an embodiment of the present invention, the object recognition unit,

Before analyzing the recognized object image, there may be provided an access monitoring system for the elderly that further includes a pre-processing unit for pre-processing the recognized object image.

In addition, according to an embodiment of the present invention, the sensor unit,

An access monitoring system for the elderly and infirm to obtain an access record of the user by monitoring the movement of the user in one direction and the other direction may be provided.

In addition, according to an embodiment of the present invention, the communication unit,

An access monitoring system for the elderly and infirm that communicates with the display unit provided by the manager in real time to provide a recognized object image and an access record of the user may be provided.

According to the access monitoring system for the elderly of the present invention, there is an effect that the guardian checks in real time whether the elderly are safe and goes out through the camera unit and the sensor unit to perform an immediate action.

In addition, it is possible to accurately recognize the elderly and infirm through the object recognition algorithm, and at the same time, it has the effect of performing an immediate action by promptly recognizing the entrant and notifying the guardian.

1 and 2 are diagrams schematically showing an elderly access monitoring system according to an embodiment of the present invention;
3 is a block diagram schematically showing a system for monitoring access for the elderly and infirm according to an embodiment of the present invention;
4 is a block diagram showing the detailed configuration of the object recognition unit of the elderly access monitoring system according to an embodiment of the present invention;
5 is a block diagram illustrating a detailed configuration of a storage server unit of an access monitoring system for the elderly and infirm according to another embodiment of the present invention.

The following objects, other objects, features and advantages of the present invention will be readily understood through the following preferred embodiments in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments described herein and may be embodied in other forms.

Rather, the embodiments introduced herein are provided so that the disclosed subject matter may be thorough and complete, and that the spirit of the present invention may be sufficiently conveyed to those skilled in the art.

Embodiments described and illustrated herein also include complementary embodiments thereof.

In this specification, the singular also includes the plural unless specifically stated otherwise in the phrase. As used herein, the terms 'comprise' and/or 'comprising' do not exclude the presence or addition of one or more other components.

Hereinafter, the present invention will be described in detail with reference to the drawings. In describing the specific embodiments below, various specific contents have been prepared to more specifically explain and help the understanding of the invention. However, those skilled in the art enough to understand the present invention can recognize that it can be used without these various specific details. In some cases, it is mentioned in advance that in describing the invention, parts that are commonly known and not largely related to the invention are not described in order to avoid confusion in describing the invention.

1 and 2 are diagrams schematically illustrating an access monitoring system for the elderly according to an embodiment of the present invention, and FIG. 3 is a block diagram schematically illustrating an access monitoring system for the elderly and infirm according to an embodiment of the present invention, FIG. 4 is a block diagram showing the detailed configuration of an object recognition unit of the elderly access monitoring system according to an embodiment of the present invention, and FIG. 5 is a detailed configuration of the storage server unit of the elderly access monitoring system according to another embodiment of the present invention. is a block diagram showing

1 to 5, the present invention, the elderly access monitoring system is largely a camera unit 100, a sensor unit 200, a storage server unit 300, a communication unit 400, and a control unit (500) configuration that includes

The camera unit 100 is configured to acquire an image of a user.

At least one camera unit 100 may be installed, and the camera unit 100 may be installed at at least one of an external entrance of a building and an internal entrance of the building.

In this case, the camera unit 100 can be installed and changed by a person skilled in the art according to the user's moving radius.

Here, the user may include the elderly, a patient, an infant, or the like.

In addition, the camera unit 100 may be various types of IP cameras, network cameras, webcams, CCTVs, etc. with built-in object recognition algorithm circuits to be described below, and the camera types include a fixed camera, a rotary camera, a dome camera, A box-type camera, a PTZ camera, etc. may be changed and used as needed, and more preferably, a vision apparatus may be used to perform precise photographing.

In addition, it is preferable that the camera unit 100 captures an image of a user, particularly an image of the elderly, to photograph the movement, movement, and the like of the elderly.

In addition, the camera unit 100 may further include an object recognition unit 110 for recognizing a plurality of objects from the user's image.

4, the object recognition unit 110 simultaneously recognizes a plurality of objects (eg, people, animals, identifiable objects, etc.) from the object image obtained from the camera unit 100, and each It is possible to receive an object image for an object, compare it with a registered image for each object previously stored in the database 140 , and determine it, and provide it to the storage server unit 300 to be described below.

Meanwhile, the object recognition unit 110 may use an object recognition algorithm such as You Only Look Once (YOLO) or Single Shot Multibox Detector (SSD) to simultaneously recognize a plurality of objects.

YOLO is an algorithm that can improve training and detection speed by approaching object recognition by regression and simplifying the overall structure. output in the form This tensor divides the image into a grid shape to express each region, and through this, the region object is recognized.

In addition, SSD is an algorithm that recognizes objects using feature maps of various sizes to generate candidate regions. The size of the feature map obtained from the CNN model decreases as the convolutional layer progresses, and the SSD All the feature maps extracted in this process are used in the inference process to recognize objects, and objects can be recognized more accurately than YOLO.

In addition, in order to increase the object recognition rate when the camera unit 100 recognizes an object, AdaBoost, Support Vector Machine (SVM), Linear Discriminant Analysis (LDA), Principal Component Analysis (Principal) Algorithms such as Component Analytic: PCA) may be further included.

In particular, when the Adaboost algorithm is used, similar objects can be recognized as the same object by analyzing the feature points of the object. It may be the same person wearing glasses, wearing a hat, or a person with a change in expression.

All of these algorithmic techniques identify the area to be recognized based on the appearance, and the area around the feature point of a face or object is detected using a model trained by a set of captured images to be used for training, and various surrounding constraints are applied. Since it is overcome through training, as a result, the accuracy and reliability of face or object feature point recognition can be increased.

In addition, the object recognition unit 110 may further include a heartbeat analyzer 120 .

The heartbeat analyzer 120 extracts a skin region from the object image transmitted from the camera unit 100 to the object recognition unit 110, and based on the value of Hue, which is the H channel, in the HSV model of the skin image. The skin area can be determined.

Here, the HSV model expresses color using three components: Hue (hue), Saturation (saturation), and Value (brightness).

In this case, when the skin image captured by the camera unit 100 is an RGB model, the heartbeat analyzer 120 may convert the skin image of the RGB model into an HSV model.

In addition, the heartbeat analyzer 120 detects a heartbeat signal through a change in the G (Green) channel in the RGB model of the object image captured by the camera unit 100, or detects a change in skin color from the skin region of the HSV model. heartbeat signal can be detected.

Here, the RGB model is the most basic color model and expresses a color as a combination of three components: Red, Green, and Blue.

In addition, the heartbeat analyzer 120 may include a band-pass filter configured as an RLC circuit to pass a specific frequency band to remove noise included in the heartbeat signal, and instead of the band-pass filter It can also be configured as a combination of a low-pass filter and a high-pass filter.

In addition, the heartbeat analyzer 120 may apply an adaptive filter such as a Kalman filter to be less sensitive to changes in skin tone or color due to changes in lighting.

In addition, the object recognition unit 110 may further include a preprocessor 130 for pre-processing the object image through correction before analyzing the object image.

Meanwhile, the pre-processing unit 130 may pre-process the object image obtained from the camera unit 100 . After extracting at least one frame from the object image data, the preprocessor 130 may adjust image scales such as pixels, colors, and sizes.

Here, an image such as movement and movement of the elderly photographed by the camera unit 100 is transmitted to the pre-processing unit 130 in real time, or an operation of correcting an image such as movement or movement of the elderly person is performed using Python or OpenCV (OpenCV). The object image transmitted to the object recognition unit 110 can be precisely recognized by performing it through a programming library such as source computer vision).

In this case, the preprocessor 130 may further include an extension conversion unit 131 , wherein the extension conversion unit 131 converts the object image captured by the camera unit 100 to the object recognition unit 110 . ) or a process of converting the extension to be displayed in the storage server unit 300 to be described below may be performed.

For example, the extension conversion unit 131 converts the object image h264 captured by the camera unit 100 into an extension such as mp4, and then the object recognition unit 110 or a storage server unit to be described below ( 300) will be transmitted.

The database 140 stores a plurality of registered images for each object in order to determine whether the object image is the same object authenticated by the object recognition unit 110 .

In addition, the database 140 stores a plurality of registered images of a specific object, that is, the elderly and infirm, and the object recognition unit 110 can increase the recognition rate of the same object through deep learning and machine learning-based learning. .

The sensor unit 200 is installed on one side of the camera unit 100 to obtain a user's access record.

The sensor unit 200 may be installed on one side of the camera unit 100 , that is, at least one of an external entrance of a building and an internal entrance of the building.

At this time, the sensor unit 200 can be installed and changed by a person skilled in the art according to the user's moving radius.

On the other hand, the sensor unit 200 may be installed in plurality, and may be a sensor capable of detecting a user, such as an Arduino infrared PIR sensor, a proximity sensor, a magnetic sensor, etc. to detect the elderly, and the proximity sensor is electromagnetic induction Any one of a high-frequency oscillation sensor that detects using , a magnetic sensor that detects using a magnet, or a capacitive sensor that detects a change in capacitance between an object and the sensor can be selected, but it is limited to the proximity sensor However, it should be noted that changes can be made by those skilled in the art as needed.

In addition, when the sensor unit 200 detects a user, the above-described camera unit 100 may be operated through the control unit 500 to be described below to obtain an image.

The sensor unit 200 may acquire the user's access record by monitoring the movement of the user in one direction and the other direction, respectively.

More specifically, the sensor unit 200 is installed at the entrance of a building, and when the user goes out of the building, one side of the sensor unit 200 detects it, and then records the user's exit from the building (user personal information, access time, etc.) can be acquired, and when the user or an outsider goes out into the building, the other side of the sensor unit 200 detects it, and then the access record indicating that the user or outsider entered the building ( User's personal information, access time, etc.) can be obtained.

That is, the sensor unit 200 is used as a means for quickly checking whether the user enters or exits and whether the user enters or exits by sensing the user's movement in one direction and the other direction, respectively.

The storage server unit 300 is configured to store the object image obtained from the camera unit 100 and the user's access record.

The storage server unit 300 is a cloud-based function in which a plurality of the storage server units 300 are distributed in a plurality of servers in connection with the camera unit 100 through communication to store object recognition and access records. can be performed quickly.

As shown in FIG. 5 , the storage server unit 300 may further include an object recognition unit 310 , a heartbeat analyzer 320 , a preprocessor 330 , and a database 340 according to another embodiment. can

The object recognition unit 310 , the heart rate analyzer 320 , the preprocessor 330 , and the database 340 have the same functions as each unit built into the camera unit 100 according to the above-described embodiment. , so a description below will be omitted.

The communication unit 400 is configured to communicate the object image stored in the storage server unit 300 and the user's access record with a display unit provided by the manager in real time, wired or wireless.

The communication unit 400 may include one or more communication modules capable of a wireless communication network, and the communication unit 400 may include a wireless communication or short-range communication module or a location information module.

For example, the wireless communication module refers to a module for wireless Internet access, and the wireless Internet module may be installed internally or externally in any one location of the inside or the outside of a building.

As wireless Internet technology, WLAN (Wireless LAN), WiFi (Wireless Fidelity), Wibro (Wireless broadband), Wimax (World interoperability for Microwave Access), HSDPA (High Speed Downlink Packet Access), etc. may be used.

In addition, the short-distance communication module refers to a module for short-distance communication, and the short-distance communication module may be installed inside or outside the building at any one location.

As the short-range communication technology, Bluetooth, Radio Frequency Identification (RFID), Infrared Data Association (IrDA), Ultra Wideband (UWB), ZigBee, WiHD, WiGig, etc. may be used.

The control unit 500 is configured to control the camera unit 100 , the sensor unit 200 , the storage server unit 300 , and the communication unit 400 .

The control unit 500 may be controlled through a control server such as a Raspberry PI, and the control unit 500 is composed of a plurality of the camera unit 100 and the sensor unit 200 and , the storage server unit 300 and the communication unit 400 may be controlled respectively.

In addition, the control unit 500 may control by applying an artificial intelligence algorithm.

More specifically, the sensor unit 200 may apply a machine learning algorithm to determine error data (whether the user enters or not) and increase the error situation recognition rate to increase the determination rate of whether the user enters or not.

More specifically, by applying a specific machine learning algorithm to the data input from the sensor unit 200 to learn the machine learning algorithm, it is possible to isolate error data that is momentarily erroneously measured due to surrounding disturbance. In this way, data processing in a form robust to instantaneous external noise and the like is possible.

In one embodiment, the machine learning algorithm may apply a binary classification method for learning and determining whether the data input from the sensor unit 200 is normal or abnormal. In addition, in the case of the above-described Binary Classification, since a lot of time is required for initial learning, there may be a disadvantage that the accuracy of the initially trained model is not high. An instantaneous error in data received from the sensor unit 200 may be appropriately handled.

In another embodiment, it is preferable to learn the user's entry/exit data into a plurality of layers by using a Convolutional Neural Network (CNN) method among deep neural network methods for classifying such data, and in particular, SVM (Supported Vector) Machine), it is desirable to classify whether the user enters or not.

In particular, in the case of SVM, by vectorizing data acquired through various sensors of the sensor unit 200 described above and learning about the vector, pattern vectors of sensor values generated when the user enters and when the user exits are generated. It is possible to effectively remove noises by effectively classifying the pattern vectors of the sensor values.

Therefore, according to the access monitoring system for the elderly of the present invention, there is an effect that the guardian checks in real time whether the elderly are safe and goes out through the camera unit and the sensor unit to perform an immediate action.

In addition, it is possible to accurately recognize the elderly and infirm through the object recognition algorithm, and at the same time, it has the effect of performing an immediate action by quickly recognizing the entrant and notifying the guardian.

The embodiments described in this specification and the configurations shown in the drawings are merely the most preferred embodiment of the present invention and do not represent all the technical spirit of the present invention, so various equivalents that can be substituted for them at the time of the present application It should be understood that there may be variations and examples.

100: camera unit 110: object recognition unit
120: heart rate analysis unit 130: pre-processing unit
131: extension conversion unit 140: database
200: sensor unit 300: storage server unit
310: object recognition unit 320: heart rate analysis unit
330: preprocessor 340: database
400: communication unit 500: control unit

Claims (6)

a camera unit for acquiring an image;
a sensor unit installed on one side of the camera unit to obtain a user's access record;
a storage server unit for storing the object image obtained from the camera unit and the user's access record;
a communication unit interworking with the storage server unit for wired/wireless communication of object images and user's access records to the manager;
a control unit for controlling the camera unit, the sensor unit, the storage server unit, and the communication unit;
including,
The control unit, the elderly access monitoring system, characterized in that controlled through an artificial intelligence algorithm.
The method according to claim 1,
The camera unit,
Recognizing a plurality of objects from the object image, the elderly access monitoring system, characterized in that it further comprises an object recognition unit for obtaining the recognized object image.
3. The method according to claim 2,
The object recognition unit,
An access monitoring system for the elderly, characterized in that the object is recognized by using at least one of the SSD algorithm, the YOLO algorithm, and the AdaBoost algorithm.
3. The method according to claim 2,
The object recognition unit,
Before analyzing the recognized object image, the elderly access monitoring system, characterized in that it further comprises a pre-processing unit for pre-processing the recognized object image.
The method according to claim 1,
The sensor unit,
The elderly access monitoring system, characterized in that the user's access record is obtained by monitoring the movement of the user in one direction and the other direction, respectively.
The method according to claim 1,
The communication unit,
The elderly access monitoring system, characterized in that communicating with the display unit provided by the manager in real time to provide the recognized object image and the user's access record.

Images