KR102284037B1 - AI Welfare Health Care Service system - Google Patents

Abstract

The present invention relates to an AI welfare health care service system, and more particularly, to an AI welfare health care service system where a dedicated application is installed on a smartphone to collect surrounding information. The collected information is immediately transmitted through text messages (SMS, LMS, MMS) even when Internet use such as Wi-Fi, 3G, 4G is unavailable or blocked, and daily life data of the elderly and emergency situations occur. So, a welfare worker or guardian (administrator) can predict a dangerous situation through the daily life data or be dispatched immediately when the situation occurs. The AI welfare health care service system includes a smartphone, a management server, a welfare worker or guardian's terminal, and a physical module.

Description

AI Welfare Health Care Service System {AI Welfare Health Care Service system}

The present invention relates to an AI welfare health care service system, and more specifically, a dedicated application is installed on a smartphone to collect surrounding information, and the collected information is unavailable to Internet use such as Wi-Fi, 3G, 4G, and Even in the blocked state, the daily life data of the elderly and emergency situations are immediately transmitted through text messages (SMS, LMS, MMS), so that the welfare worker or guardian (administrator) can predict a dangerous situation through daily life data, or be immediately dispatched when a situation occurs. It is about an AI welfare health care service system that provides

Looking at the conventional IOT service and problems, the installation, management, and operation of the application was very complicated due to the complex WCDMA, LTE, and 5G system configuration.

One manager who provides welfare health care services took care of an average of 143 people, and over 4 years, there were 13,000 malfunctions.

As a result, maintenance costs increased due to smartphone malfunction, and in-person visits, surveys, and home care tasks were aggravated.

In 2018, the stress caused by the increase in work such as inspection and report preparation of national IOT safety service equipment and dispatching to equipment malfunctioning increased, and the number of firefighting dispatch requests due to sensitive operation increased by 38%.

In addition, there was a problem in that the existing IOT devices that provide elderly care must have internet (Wi-fi) to provide continuous service as a safe service for the elderly living alone.

Korean Patent No. 2077594 Korea Registered Patent No. 1997225 Korean Patent No. 1721853 US Patent No. 10143415

The present invention has been made to solve the above problems, and an object of the present invention is to provide an AI welfare healthcare service system that can be operated without the Internet (data, Wi-fi, etc.).

In addition, the present invention uses only the basic functions within the smartphone that do not require an external sensor or input device, so to provide an AI welfare healthcare service system that provides a service that does not incur application costs for automatic message transmission. There is a purpose.

Another object of the present invention is to provide an AI welfare healthcare service system connected to a smart watch, a smart mirror, a smart robot, a smart refrigerator, and an AI speaker through a network.

In order to solve the above problems, the present invention is a smartphone in which the security service application is installed and the elderly or protected persons possess; The safe service application can automatically input and send text messages in regular/emergency situations through the transmission function of the text message disclosure API, and the personal information, phone number, phone number, and daily life information of the elderly living alone or the person to be protected can be stored in the database. a management server that manages the system; and a terminal of a welfare worker or guardian capable of voice or video call with the smart phone of the elderly living alone and the person to be protected.

The smart phone collects data on whether the smartphone moves faster than a certain speed through the speed sensor, whether an impact occurs or whether a fall occurs through the shock sensor, and sends an alarm when it exceeds a certain number or number of times, so that the elderly living alone or the person to be protected sends an alarm Check if the alarm is turned off, and if the alarm is not turned off, it is recognized as an abnormal symptom and an emergency notification is sent to the management server via text message.

The management server sends a text message to the smartphone to determine whether the smartphone is not used for a certain period of time or whether it is alive or dead, the application installed on the smartphone receiving the text message identifies the received contact, and the management server When a text message comes to a contact identified as Send a text message to notify you that there is an error.

It further includes; a hidden module that sends the automatically collected information to SMS/LMS/MMS regularly/irregularly without manually writing the text message in the smartphone.

The smart phone checks a physical phenomenon obtained from the acceleration sensor and the impact sensor, and a physical module capable of sending it to SMS/LMS/MMS; further includes.

The hidden module controls so that the service is kept active in the background state without cumbersomely turning on the application again even when the smartphone is turned off and on, or the application is automatically terminated.

If it is necessary to check whether the smartphone is turned off, the management server sends a text message for outgoing to another smartphone that needs confirmation, receives the message normally from the smartphone, sends the received message again, and sends a reply to the server by replying to the corresponding smartphone It includes an automatic text message sending module that can confirm that this is operating normally.

The security service application of the smartphone transmits the information to the management server when the smartphone has no or low call volume, and uses the artificial intelligence facial recognition model or algorithm for the elderly living alone or the person to be protected using the smartphone and an artificial intelligence service module that provides a service that can reduce loneliness and depression by making a voice call from the management server to the corresponding smart phone based on the analyzed information.

The present invention made as described above can be operated without a separate data usage fee even when there is no internet (data communication, Wi-fi, etc.) or when it is difficult to operate due to a natural disaster.

In addition, the present invention can analyze automatically in the background state even if it is not turned on by simply installing the App on the smartphone, and accordingly, it is possible to provide a continuous service for the elderly living alone, as well as to facilitate the expansion of new services.

In addition, the present invention records the data of the elderly living alone or the person to be protected in the application simply by installing the application, and the data is automatically sent every day through a text message, and the text message is not used by the user using early morning hours. It is possible to reduce the reception error when data is sent at the same time by sending it at the right time or randomly sending at a specific time in order to reduce the reception error of text messages sent by many people at once.

In addition, when a new device such as a smart watch, smart mirror, smart robot, smart refrigerator, AI speaker, etc. is connected, data information (health information) obtained from the new device is transferred from the smartphone to the Internet while Bluetooth is communicating with the smartphone. In the future, the company's service can be expanded to send text messages to the server instead of sending them to the server.

In addition, since the present invention uses only basic functions in the smartphone, there is no additional cost, and the service can be provided at a low cost.

1 is a view showing the problems of the aged care system according to the prior art.
2 is a diagram showing a schematic configuration of an AI welfare healthcare service system according to an embodiment of the present invention.
3 is a diagram illustrating the effect and embodiment of the AI welfare health care service system according to an embodiment of the present invention.
4 is a flowchart showing a service step according to another embodiment of the present invention.

In order to fully understand the present invention, preferred embodiments of the present invention will be described with reference to the accompanying drawings. Embodiments of the present invention may be modified in various forms, and the scope of the present invention should not be construed as being limited to the embodiments described in detail below. This example is provided to more completely explain the present invention to those of ordinary skill in the art. Accordingly, the shape of elements in the drawings may be exaggerated to emphasize a clearer description. It should be noted that the same members in each drawing are sometimes shown with the same reference numerals. In addition, detailed descriptions of well-known functions and configurations that are determined to unnecessarily obscure the gist of the present invention will be omitted.

As shown in Figure 2, it is composed of a smartphone 11 of the elderly living alone or a person to be protected, a terminal 12 of a welfare worker or guardian, a management server 30, and the like.

As an embodiment, the present invention provides a secure service application installed, the smart phone 11 possessed by an elderly person living alone or a person to be protected, and the safe service application automatically transmits a text message in case of a regular/emergency situation with the transmission function of the text message disclosure API. A management server 30 that can input and send, and manages the personal information, phone number, phone number, phone number, and daily life information of the elderly living alone or the person to be protected into a database, the smart phone 11 of the elderly living alone and the person to be protected; It includes a terminal 12 of a welfare worker or guardian capable of voice or video call with the smartphone 11 .

When the application is installed on the smart phone 11 of the elderly living alone or the person to be protected, member information is entered, and the guardian code received through the terminal 22 of the welfare worker or guardian is transmitted to the management server 30, management It receives a text from the server 30 indicating that the input is complete and completes it.

For example, the security service application 12 of the smartphone 11 is (1) a service method that operates in the background so that the service can be operated without the elderly living alone or the person being protected using the application or intentionally activating the application, and the smart The phone 11 collects data on whether or not the phone 11 moves faster than a certain speed through the speed sensor or whether an impact or fall occurs through the shock sensor, and sends an alarm if the number or number of falls is greater than a certain number or number of times to alert the elderly living alone or the person being protected is turned off, and if the alarm is not turned off, it can be recognized as an abnormal symptom and urgently notified to the management server 30 through a text message.

Alternatively, the application installed on the smartphone 11 that does not use the smartphone 11 for a certain period of time, or sends a text message from the management server to the corresponding smartphone 11 in order to determine whether the smartphone 11 is alive or dead, and receives the text message. This received contact is identified, and when a text message comes to the contact identified by the management server, the application exposes a notification to the elderly living alone or the person to be protected. , and if the notification is not turned off, the application can send a text message to the server again to notify that there is something wrong.

It operates as a background program of the smart phone 11 in advance of an application including a message transmission and reception function, which is an existing application installed in the smart phone 11, and collects data list for activity check as collection information, wake up /Bedtime, smartphone use, walking distance, phone calls, battery information, shock/fall and condition abnormality information, smartphone information, name, gender, date of birth, phone number, residence (activity area), place of activity ( home, welfare center, etc.), disease history, and contact information of guardians.

Prior to the safe service application, while operating in the background state of the smart phone 11, the message transmitted and received through the safe service application is recognized and managed.

The terminal 22 of the welfare worker or guardian issues a guardian code (eg, 00 - 0 - 00 (8 digits)) as one of the welfare health care services and transmits it to the smart phone 11 of the elderly living alone or the person to be protected.

As shown in Fig. 2 (b), when member information is changed through the smart phone 11 of the elderly living alone or the person to be protected, a text message of the member information change data is sent to the management server 30, or a daily text message (message for behavioral pattern data analysis, etc.) to the management server 30, or transmits the information to the management server 30 when an emergency (eg, battery / fall, etc.) occurs by various sensor information.

The management server 30 stores the member information of the elderly living alone or the person to be protected, receives the member behavior pattern input through the smartphone 11, collects and predicts the transmitted data, and when abnormal symptoms occur, the terminal of the welfare worker or guardian ( 22), and provides information so that those with authority can read the records.

In addition, the management server 30 sends a daily report to the terminal 22 of the welfare worker or guardian, or a search service for behavior analysis data collected through the smartphone 11, a message notification service in case of an emergency, and member information A confirmation message is sent when data is received successfully when changing.

In addition, the management server 30 can systematically manage statistics and records as a statistical/girol management tool, and can automatically notify the situation occurrence to the 119 fire station server 41 .

Specifically, a hidden module for sending the automatically collected information regularly/irregularly to SMS/LMS/MMS without having to manually write a text message in the smartphone 11; further comprising; The module controls so that the service is kept active in the background state without cumbersomely turning on the application again even when the smartphone 11 is turned off and on, or the application is automatically terminated.

a physical module that allows the smart phone 11 to check a physical phenomenon obtained from an acceleration sensor and an impact sensor and send it to SMS/LMS/MMS; When it is necessary to check whether the smart phone 11 is turned off, the management server sends a text message for outgoing to the other smart phone 11 that needs confirmation, and receives the message normally from the smart phone 11 and sends the received message again An automatic text message sending module that can confirm that the corresponding smart phone 11 is operating normally by sending and replying to the server; And the security service application of the smartphone 11, when the smartphone 11 has no or low call volume, transmits the information to the management server, and the elderly or protected persons using the smartphone 11 An artificial intelligence service module that provides a service that can reduce loneliness and depression by making a voice call from the management server to the corresponding smartphone 11 based on the information analyzed using the artificial intelligence facial recognition model or algorithm for further includes ;

As shown in FIG. 3 , the management server 30 may transmit a message regarding the welfare of the 'old person living alone or the name of the person to be protected' to the smartphone 11 .

For example, in the area of 'the elderly living alone or the person being cared for', today's weather (overcast and occasional rain, maximum 16℃ / minimum 1℃, fine dust: good 23 μg/㎥) briefing and activity briefing" Yesterday, early in the morning After waking up, you walked 3Km and moved 20Km by car! (Activity 5% ↑ compared to the previous day)" is sent to the smartphone (11) on a regular basis.

In addition, the management server 30 further includes an AI (artificial intelligence) healthcare diagnostic module, and by the information measured by the smart phone 11 or the wearable device described below, " 1) You have high blood pressure, so it is a cold and rainy day. It is dangerous to stay outside for a long time!Please call to say hello and refrain from going out. 2) The number of open smartphones (11) is high compared to the number of calls, 3) Low activity compared to the average age (recommended to walk 3 km more), 3) The amount of activity has steadily decreased over the past month, do you not have any health problems?" and transmits information such as "to the smart phone 11 and the terminal 22 of the welfare worker or guardian.

In addition, the management server 30 through the detailed data analysis module, the analysis date (March 26, 2019), the expected wake up time (6:30 am), the expected bedtime (1:40 pm), the smartphone opened Detailed data such as number of times (105 times), number of calls (6 times), walking distance (3Km), and vehicle travel distance (20Km) are transmitted to the terminal 22 of the welfare worker or guardian to enable detailed management (Fig. Please refer to 2, Call for Confirmation/Condolences).

Specifically, as a text message sent every day, the text message information sent daily (up to 90 bytes) by SMS from the smartphone 11 of the elderly living alone or the person being protected to the server can be checked with 1) mobile phone (sender) number / calling contact number. , 2) the number of times the smartphone was opened, 3) the first/last smartphone use time, 4) the walking/vehicle travel distance (Km) using the GPS module, and 5) the number of calls.

Regular text messages include text messages (up to 90 bytes) that are sent immediately when an event occurs, 1) less than 5% battery, and 2) a response request message when there is no response for 3 minutes after a fall notification.

As shown in Fig. 4 (a), when the guardian presses the send button to the terminal 22 of the welfare worker or guardian after modifying the information, it is sent as a text message (transmitted to LMS; up to 2,000 bytes) (S101 ~ S104).

The text message includes 1) name (up to 4 characters), 2) gender, 3) date of birth, 4) mobile phone (caller) number, 5) address, 6) activity location (maximum of 2) / maximum of 10 characters, 7 ) disease history (up to 5) / maximum of 7 characters, 8) contact information of guardian (up to 3) / name of guardian (up to 5 characters, contact number 11 characters, 9) guardian code number / 8 digits, etc. are included.

As shown in Fig. 4 (b), the process (S201 ~ S204) is shown when data does not come to the server or when data of the past date is desired to be transmitted.

First, the management server sends a text message for confirmation to the smartphone 11, receives the text from the smartphone 11, and sends one more time if the text message cannot be returned from the smartphone 11 to the management server, or management You can recheck the reply message from the server.

Or, when a text message cannot be returned to the management server, a notification text is sent to the guardians (terminals).

The management server stores the OK text message reply only when the confirmation text message request is activated (the management server automatically activates data entry within 10 minutes after sending).

At this time, the reason that the management server activates data input within 10 minutes after automatic transmission is to prevent abusing and randomly sending a signal to the server.

As shown in FIG. 4 ( c ), when you want to receive data of the past date, a message for sending data is transmitted from the manager page of the management server to the corresponding smartphone 11 , and a text message is received from the smartphone 11 . After doing this, a text is sent from the smartphone 11 to the management server, and a reply message can be checked from the management server.

At this time, if there is no reply to the text message to the management server, a warning window is formed on the manager page, and if there is no reply, the server can send a message once more.

When a data is requested to the smart phone 11 on the manager page, the data stored in the smart phone 11 is programmed to be transmitted to the management server.

For reference, if the smartphone data record storage limit is 30 days, only data will be recorded and sent within 30 days, and a re-request number exists in front of the re-requested message.

On the other hand, in the above rules for generating the guardian code, 8 digits of the guardian code are given after the completion of individual/group membership registration. For example, 1) 128 - 12 - 254, 2) 248 - 45 - 849, and so on.

This is because when sending personal information after downloading a smartphone application, it is to prevent a person who has not registered as a member from randomly entering data into the server.

That is, if the third digit is not 8, even if data is input to the server, it is automatically classified as unconfirmed data.

In addition, the daily data transmission text time can be set with the combination of the 5th and 6th numbers. For example, a member managed by a guardian with a code number of 128-12-54 (1) the data transmission time specified by the company (2 a.m. hour), (2) 5th digit 2 / 6th digit 5 = 25 minutes, and this member sends a data text message at 2:25 a.m.

This is because, if the number of members increases, it is to prevent (prevent data loss) from receiving too many text messages at a specified time.

Therefore, after installing the customer application on the smartphone and entering information after entering the code number, the code number is stored in the application so that a message can be sent at the specified time.

Example

A wearable device owned by an elderly person living alone or a person to be protected includes a terminal 22 of a worker or guardian connected to a communication network network, for example, a smart watch is composed of an SOS button, a POWER button, a bezel part, and a display left and right scroll part, ,

The SOS button transmits the SOS to the terminal of the welfare worker or guardian when the SOS button is continuously pressed a certain number of times while the main screen of the wearable device is turned on,

The bezel part has a screen scrolling function, moving an item, and changing a setting value, and rotating the bezel part can scroll the screen, move an item, change a setting value such as a volume,

The display left and right scroll unit provides a response system for checking the use of a smartphone by scrolling the screen left and right to select a menu.

For example, the application installed on the smartphone that received the text message identifies the received contact, and when a text message comes to the contact identified as the management server, the application exposes a notification to the elderly living alone or the person to be protected. When the display is turned on or off by scrolling the screen left and right, the notification is automatically turned off.

If you swipe the screen to the left on the wearable device main screen to move to the VOICE screen, you can check the list of recent calls.

When a call is received, the voice reception pop-up screen is displayed, and the call can be started by pressing the call button.

The biometric information automatically measured by the elderly living alone or the person to be protected by the wearable device includes a heart rate, calories burned, and walking distance.

In addition, the face recognition model used by the camera connected to the management server 30 through a network is calculated through Equation 1 below.

는 영상의 좌표이고,(where N is the number of facial recognition pattern images used for training, M is the number of intersections found in each image,

is the coordinates of the image,

는 왜곡 모델로 추정한 좌표로써, A는 카메라의 초점 거리, 주점, 비틀림 계수, 접선 왜곡을 포함하는 벡터이고,

는 방사 왜곡 계수이며,

는 각각 i번째 영상의 외부 파라미터로써 회전과 이동을 나타내는 벡터이고,

는 j번째 좌표의 실제 3차원 좌표이다. )

is the coordinates estimated by the distortion model, A is a vector including the focal length of the camera, the main point, the torsion coefficient, and the tangential distortion,

is the radiation distortion coefficient,

are vectors representing rotation and movement as external parameters of the i-th image, respectively,

is the actual 3D coordinate of the j-th coordinate. )

Meanwhile, in the image analysis method according to an embodiment of the present invention, the management server 30 may use a specific algorithm (AdaBoost algorithm) to detect the current state of an elderly person living alone or a person to be protected.

The algorithm combines simple but fast weak classifiers to form a single strong classifier.

In accordance with an embodiment of the present invention, using a weak classifier to recognize walking samples for the elderly living alone or to be cared for, and to reduce the weight for the correctly classified elderly living alone or caretaker walking sample, misclassified elderly living alone or caretaker walking sample Finally, one strong classifier is composed of a linear combination of weak classifiers because the next classifier supplements the ones misclassified by the previous classifier by increasing the weight and reflecting it in the next weak classifier. do.

는 독거노인 또는 보호 대상자의 이동(도보, 낙상 등)영상이며,

독거노인 또는 보호 대상자 및 배경 영상으로 각각 +1 또는 -1의 값을 가진다. Table 1 shows the steps of constructing a strong classifier by linearly combining weak classifiers using an algorithm.

is a moving image (walking, falling, etc.) of an elderly person living alone or a person to be protected,

It has a value of +1 or -1, respectively, as a background image for the elderly or caretaker living alone.

가중치 오차를 최소화하는 약한 분류기

를 생성하여 각 분류기에 대한 데이터 가중치를 갱신한다. And

Weak classifier that minimizes weight error

to update the data weights for each classifier.

는 모든 데이터

에 대하여 수식을 정규화한다.

is all data

Normalize the formula against .

는 최종적으로 생성된 강한 분류기이다. 강한 분류기는 벡터

를 입력으로 받으며 훈련된 약한 분류기

의 합으로 분류를 수행하게 된다.

is the finally generated strong classifier. A strong classifier is a vector

Weak classifier trained with as input

Classification is performed by the sum of

The local binary pattern feature used to detect the elderly living alone or the person to be cared for is used to express local texture features of the face such as edges, points, and corners using pixel values. It has the advantage of being robust to changes in

The local binary pattern feature is expressed as in Equation 2, and the difference between the pixel value of the current position and the neighboring pixel value is expressed as values of 0 and 1.

와

는 중심화소의 화소값 및 이웃화소의 화소값을 의미한다.In Equation 2, P and R mean the number of adjacent pixels and the radius of the circle, respectively,

Wow

denotes a pixel value of a central pixel and a pixel value of a neighboring pixel.

For example, in the case of LBP(8, 1), 8-bit encoding of each pixel can be obtained by comparing the pixel values of 8 neighboring pixels based on the central pixel value and assigning 1 if the neighboring pixel value is larger and 0 if smaller. .

The processor of the camera of the present invention uses a probabilistic data association filter to track the detected elderly living alone or caregiver object.

Here, the correlation of data extracted from the smartphone 11 refers to a method of estimating the source of a measurement value obtained in a clutter environment in which ambiguity about the source of the measurement value exists.

The probabilistic data association filter calculates the probability of being associated with the trajectory of the next elderly living alone or the object to be protected after obtaining the measured values such as the movement speed and location of the object from the detected elderly living alone or the object to be cared for, It is a technique to obtain an estimate of the trajectory by summing it up.

The probabilistic data association filter basically follows the structure of the Kalman filter, and largely consists of a prediction step, a step of confirming the validity of the measured value, a step of confirming the correlation of the data, and finally a step of estimating the state.

For example, if the camera or probabilistic data-related filter is moving faster than a certain speed or if the shock sensor collects data on whether an impact or fall occurs, and the amount of impact is greater than a certain value or the number of falls is greater than a certain value By sending an alarm, it is checked whether the elderly living alone or the person to be protected turns off the alarm.

A convolutional neural network is used to classify the tracked elderly or caretaker objects by gender. Convolutional neural networks are a type of multi-layer perceptrons designed to minimize preprocessing.

Like other neural networks, it consists of an input layer, a hidden layer, and an output layer, and each layer consists of one or more neural networks.

The hidden layer consists of one or several convolutional layers and a pooling layer, and has the advantage of being able to fully utilize the input data of a two-dimensional structure, and thus is widely used in the image recognition field.

As a method of training data with a convolutional neural network, a back propagation algorithm is used.

The back-propagation algorithm may be composed of a feed forward and a back-propagation step. In the feedforward stage, the training data is applied to the neural network to obtain the error and cost function from the output stage.

Here, while the calculated error value is propagated back, the weight and bias values are updated by gradient descent, and the process is repeatedly performed so that the weights and bias values become optimized values.

Therefore, the present invention uses an algorithm to detect the elderly living alone or the person to be cared for through the camera, and each part of the elderly living alone or the person to be cared for is respectively a Har-like feature and a local binary pattern (LBP, local bit pattern). Train using features.

A region of interest (ROI) is detected by performing a cascade clasifier using the training result.

In addition, the artificial intelligence service module, which provides a service that can reduce loneliness and depression by making a voice call with the smartphone 11, converts the corresponding text content into an audio file by referring to the face recognition information and the current image. can be transmitted

11: Smartphone
12: Relief service application
22: terminal of the welfare worker or guardian
30: management server

Claims (8)

A smartphone owned by an elderly person living alone or a person being protected with a safety service application installed; The safe service application can automatically input and send text messages in regular/emergency situations through the transmission function of the text message disclosure API, and the personal information, phone number, phone number, and daily life information of the elderly living alone or the person to be protected can be stored in the database. a management server that manages the system; a terminal of a welfare worker or guardian capable of voice or video call with the smart phone of the elderly living alone and the person to be protected; A physical module that the smart phone can send to SMS/LMS/MMS by checking the physical phenomenon obtained from the acceleration sensor and the impact sensor;
The management server sends a text message to the smartphone to determine whether the smartphone is not used for a certain period of time or whether it is alive or dead, and an application installed on the smartphone receiving the text message identifies the received contact, and the management server When a text message comes to a contact identified as Send a text message to the server to notify you that there is an error,
The security service application of the smartphone transmits the information to the management server when the smartphone has no or low call volume, and an artificial intelligence facial recognition model or algorithm (AdaBoost) for the elderly living alone or the person to be protected using the smartphone algorithm) to provide a service that can reduce loneliness and depression by making a voice call from the management server to the smartphone through the artificial intelligence service module,
The local binary pattern used to detect an elderly person living alone or a person to be cared for is expressed as the following equation, and is used to detect local texture features of the face including edges, points, and corners using pixel values AI, characterized in that Welfare healthcare service system.
[Equation]

(Here, the difference between the pixel value of the current position and the neighboring pixel value is expressed as values of 0 and 1, and P and R mean the number of adjacent pixels and the radius of the circle, respectively,

Wow

denotes the pixel value of the central pixel and the pixel value of the neighboring pixel.) According to claim 1,
It collects data on whether the smartphone moves faster than a certain speed through the speed sensor or whether an impact or falls have occurred through the impact sensor, and sends an alarm when the amount of impact is greater than a certain value or the number of falls is greater than a certain value. AI welfare healthcare service system, characterized in that it checks whether the elderly or the person to be protected turns off the alarm, and if the alarm is not turned off, it is recognized as an abnormal symptom and urgently notified to the management server by text message. delete According to claim 1,
AI welfare health care service system, characterized in that it further comprises; a hidden module for sending the information automatically collected from the smart phone included in SMS / LMS / MMS regularly or irregularly. delete 5. The method of claim 4,
AI welfare healthcare, characterized in that the hidden module controls the application so that the service is activated in the background state without turning the smartphone off and on, or turning the application back on in a cumbersome state even when the application is automatically terminated service system. According to claim 1,
When it is necessary to check whether the smartphone is turned off, the management server sends a text message for outgoing to another smartphone that needs confirmation, receives the message normally from the smartphone, sends a received message again, and sends a reply to the management server to reply to the corresponding smart phone. AI welfare healthcare service system comprising a; automatic text message sending module that can confirm that the phone is operating normally. delete

Images