KR101259560B1 - Apparatus and method for fall-down detection, and system for emergency aid using the same - Google Patents

Abstract

The present invention relates to a fall detection device and method for actively preparing for a second accident caused by a fall by detecting a fall accident frequently occurring in real life for an elderly person or a pedestrian with inconvenient mobility. In an embodiment, the fall detection apparatus may include an acceleration sensor that detects acceleration and inclination information of a user, and when a fall is determined by the determination module and the determination module that determine whether a fall is made using the acceleration and inclination information extracted from the acceleration sensor, the fall occurrence information. It consists of a wireless communication module for transmitting.

Description

Wrist-wearing fall detection device and method and fall detection system using same {APPARATUS AND METHOD FOR FALL-DOWN DETECTION, AND SYSTEM FOR EMERGENCY AID USING THE SAME}

The present invention relates to a wrist worn device having a three-axis acceleration inertial sensor and a method thereof and an emergency rescue method using the same. It relates to a fall prevention technique and a behavior pattern recognition technique to transmit.

The present invention is derived from research carried out as part of the IT original technology development project of the Ministry of Knowledge Economy. [Task Management Number: 2008-F039-01 Title of Task: Development of Human-Robot Interaction Mediation Technology]

Along with the development of IT technology, it is possible to remotely visit or consult in places such as general homes and nursing homes, not in special places or facilities such as medical institutions. Especially for chronically ill and elderly people, constant observation is most important.

Older people are more likely to fall into emergencies, and they are less able to cope with emergencies and can cause serious situations if they do not take appropriate measures. In particular, falls can be caused by a number of diseases and can be caused by simple aging, requiring immediate response in case of a precaution or situation.

The prior art detects a fall from an inertial sensor and transmits emergency information when a fall occurs using a miniaturized device worn on the waist or chest. However, there are many false positives that are not falls, and there are many false negatives that turn out to be fall despite the fall, which wastes unnecessary time and manpower.

The present invention was conceived by recognizing the need to provide a fall detection device that does not restrain the user and an emergency rescue service using the same, the problem to be solved by the present invention is the fall of any operation of the user's daily hands and arms In order not to be determined, the present invention provides a wrist wearable fall detection device and a method attached to a hand and an arm that can easily extract a shock or a change in behavior with the ground.

Another problem to be solved by the present invention is to provide an emergency rescue service for transmitting an emergency call to a corresponding institution when an actual emergency occurs using a wrist worn fall detection device.

The objects of the present invention are not limited to the above-mentioned objects, and other objects not mentioned can be clearly understood by those skilled in the art from the following description.

Falling detection device according to an aspect of the present invention for achieving the above object is a sensor module for collecting the acceleration and inclination information, and a determination module and determination module for determining whether or not the fall using the acceleration and inclination information collected from the sensor unit It includes a communication unit for transmitting the fall information determined from the external emergency response agency.

According to another aspect of the present invention, a fall detection method includes: collecting acceleration and slope information, determining whether a fall is made using the detected acceleration and slope information, and using the fall information to determine whether the fall information is external to an emergency situation. Transmitting to the coping organization.

The fall rescue method according to another aspect of the present invention comprises the steps of receiving the fall information from the fall detection device and the first notification of the occurrence of the fall to the external emergency-related organizations, the static section detection performed in the fall detection device Receiving a result and finally notifying the emergency-related organizations of the static section detection results.

In the case of using the fall detection device according to the present invention, without restraint and inconvenience to the user, it is possible to distinguish between the normal operation and the fall can accurately detect the fall, when the fall occurs in everyday life, the first emergency call As the analysis is divided into the second emergency call and the second emergency call, there is an advantage in that the reliability of the final fall judgment and the emergency judgment can be improved.

1 is a conceptual diagram of a fall detection system according to an embodiment of the present invention.
Figure 2 is a block diagram of a wrist worn fall detection device according to an embodiment of the present invention.
3 is an internal configuration diagram of a determination module constituting a wrist worn fall detection apparatus according to an embodiment of the present invention.
4 is a flowchart illustrating a fall detection method according to another embodiment of the present invention.
5 is a flowchart illustrating a NHC calculation method of the fall detection method according to an embodiment of the present invention.
6 is a flowchart illustrating a NEC calculation method of the fall detection method according to an embodiment of the present invention.

Advantages and features of the present invention, and methods of achieving the same will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. The present invention may, however, be embodied in many different forms and should not be construed as being limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. Is provided to fully convey the scope of the invention to those skilled in the art, and the invention is only defined by the scope of the claims. It is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. In the present specification, the singular form includes plural forms unless otherwise specified in the specification.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the drawings, the same reference numerals are used to designate the same or similar components throughout the drawings. In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear.

A fall detection system using a wrist worn fall detection apparatus according to an exemplary embodiment of the present invention will be described with reference to FIG. 1. 1 is a conceptual diagram of a fall detection system according to an embodiment of the present invention.

Referring to FIG. 1, a fall detection system according to an embodiment includes a wrist worn fall detection apparatus 100, an emergency call gateway 200, and an emergency response related organization 300.

The wrist wearable fall detection device 100 determines whether a fall occurs by using information such as acceleration and inclination according to a user's movement, and when it is determined that a fall occurs, the wrist wearable fall detection device 100 transmits the fall occurrence information through wireless communication to the emergency call gateway 200. Send to). Here, the fall detection apparatus 100 may transmit the fall information to the emergency call gateway 200 using a wireless communication method including Zigbee and Bluetooth.

The emergency call gateway 200 includes a wireless receiving module 201 for receiving the falling information transmitted from the fall detection device 100 and a transmitting unit 202 for transmitting the falling information to the emergency response-related organ 300. . Here, the emergency call gateway 200 may transmit the falling information to the emergency response agency 300 using a public network / wired communication protocol including the Internet, PSTN, CDMA.

Referring to Figures 2 and 3 will be described wrist worn fall detection device according to an embodiment of the present invention. 2 is a block diagram of a wrist worn fall detection apparatus according to an embodiment of the present invention, Figure 3 is an internal configuration of the determination module constituting a wrist worn fall detection apparatus according to an embodiment of the present invention.

2, the fall detection apparatus 100 according to an embodiment includes an acceleration sensor 110, a determination module 120, and a wireless communication module 130.

The acceleration sensor 110 may use various acceleration sensors capable of detecting acceleration information and tilt information. In the present invention, a three-axis acceleration sensor is described as an embodiment. The 3-axis acceleration sensor collects 3-axis acceleration information and tilt information according to the user's movement.

The determination module 120 determines whether a fall action occurs using the acceleration information and the slope information collected from the acceleration sensor 110, and transmits the fall information to the emergency call gateway 200 through the wireless communication module 130. do.

Referring to FIG. 3, the determination module 120 includes a preprocessor 121, a Norm calculator 122, a NHC (Norm Hit Crossing Count) calculator 123, a NEC (Norm Extremum Count) calculator 124, and The determination unit 125 is included.

The preprocessor 121 is configured to receive the acceleration information and the slope information from the acceleration sensor 110, and is composed of a filter for removing unnecessary noise included in the signals, the fall of the noise after removing the noise from the signals The frame is extracted and stored in a size that can include a section. Here, the preprocessing unit 121 repeatedly processes the signals by sampling the signals to form a frame, and overlapping a frame in which a fall occurs in the same or smaller frame than the previous frame.

The Norm calculator 122 extracts Norm, that is, the magnitude component of the entire 3-axis acceleration, of the vector to be used to extract the four features to be used in the determination module 120. In addition, the Norm calculator 122 may include a low pass filter, and pass the extracted Norm through the low pass filter to extract a static norm signal having a reduced dynamic component.

In particular, the static norm signal may be useful for distinguishing the static motion from the dynamic motion in acceleration information or inclination information having a lot of impact or motion components.

The NHC calculator 123 extracts a Norm Hit Crossing Count (hereinafter referred to as NHC) based on the Norm of the vector extracted by the Norm calculator 122. Specifically, the NHC calculator 123 finds the maximum range of the amplitude of the vertical axis, that is, Norm, in the given time interval frame, subdivides the interval from the lowest to the highest interval so that the maximum range can be scanned at a given resolution, and the first The bias point is moved from the interval point to the biased zero-crossing points, and the result is repeated up to the maximum value interval. When one frame is repeated, a distribution of large and small values according to the interval of the magnitude component of Norm can be obtained from the vertical axis, that is, the amplitude point of view. Next, the NHC calculator 123 converts the calculated value into a sampling frequency such as Norm so that the calculated value can be compared with the Norm value.

The NEC calculation unit 124 extracts a Norm Extremum Count (hereinafter referred to as NEC) indicating how much Norm vibrates along the time axis in terms of the horizontal axis, that is, the time axis in one frame of the Norm, and the Euclidean between NEC and Norm. Calculate the distance.

Unlike the NHC calculator 123, the NEC calculator 124 divides a given frame into subframes along a time axis, buffers the subframe sections, and localizes the maximum value of each subframe section, Extract the minimum values and store the count value of the corresponding subframe.

Next, the NEC calculator 124 signals the count value of the stored subframe, calculates and signals the Euclidean distance between NEC and Norm so that this data is a valid value.

The determination unit 125 includes the Norm and Static Norm extracted from the Norm calculator 122 and the Euclidean distance signal extracted from the NHC and NEC calculator 124 extracted from the NHC calculator 123 as feature vectors. By determining whether the fall, and if it is determined that the fall has occurred, the fall information is transmitted to the emergency call gateway 200.

In detail, the determination unit 125 includes a general section detection unit 126 and a static section detection unit 127. The general section detector 126 compares the Norm and Static Norm extracted from the Norm calculator 122 and NHC values extracted from the NHC calculator 123 with threshold values th1, th2, and th3 obtained through an iterative experiment. In the Euclidean distance signal extracted by the NEC calculation unit 124, it is determined whether or not a fall occurs by comparing the distance value at the place where the Norm of the maximum peak value occurs with the minimum value th4 in the section in the surrounding frame.

The general section detecting unit 126 recognizes the fall as a similar fall when three or more are determined to be true as a result of comparing the magnitudes and threshold values of the four feature vectors, and transmits the first emergency call to the emergency call gateway 200. At the same time, an operation command is given to the static section detection unit 127.

The static section detecting unit 127 does not operate in a daily living environment, and when a fall occurs, it operates when a situation is difficult because the user is injured so much that the user cannot recover on his own.

The static section detection unit 127 receives an operation command from the general section detection unit 126 to determine whether it is difficult for the user to return to daily life, and to transmit a second emergency call, which is the final decision, In a certain time range, for example, a static section detection is performed in a range of about 1 to 5 minutes. Here, the static section detection, unlike detecting only the fall in the normal section, in addition to the detection of the fall and detects whether the user's motion is static.

Specifically, during the static period (time range of 1 minute to 5 minutes after the first emergency call is transmitted), the Norm calculator 122 calculates the Norm of the vector, that is, the magnitude component of the entire 3-axis acceleration, and the extracted Norm. Is passed through the low pass filter to extract the static norm signal with reduced dynamic motion components.

The static section detecting unit 127 compares the secondly extracted Norm and Static Norm with the experimentally determined thresholds th5 and th6, and if the comparison result shows that both Norm and Static Norm have a smaller value than the threshold, the user almost operates. It is determined that not to be finally recognized as a fall and transmits the second emergency call to the emergency call gateway (200).

  The fall detection apparatus 100 may further include a location information extractor 140 to transmit the location information of the user to the emergency call gateway 200 together with the fall information.

In addition, the fall detection apparatus 100 may further include an emergency call button 150, and may manually transmit a final secondary emergency call to the emergency call gateway 200 by a user's operation.

In addition, the fall detection apparatus 100 may further include a wear detection sensor 160 to monitor whether or not the user wears the fall detection apparatus 100 at all times.

4, a fall detection method according to another embodiment of the present invention will be described. 4 is a flowchart illustrating a fall detection method according to another embodiment of the present invention. The flowchart shown in FIG. 4 is only one example, and the present invention is not limited to the time-series order shown in FIG.

First, three-axis acceleration information and slope information are collected according to the user's movement (S401). Next, unnecessary noise is removed from the collected acceleration information and the slope information so that characteristic information necessary for determining whether to fall is extracted (S402), and the frame is extracted and stored to a size that may include a falling section (S403). . Here, the preprocessing unit 121 repeatedly processes the signals by sampling the signals to form a frame, and overlapping a frame in which a fall occurs in the same or smaller frame than the previous frame.

Next, the Norm calculator 122 extracts the Norm of the vector, that is, the magnitude component of the entire 3-axis acceleration, and extracts the Norm signal into the low pass filter to extract the four features to be used in the determination module 120. Pass through to extract the static Norm signal is reduced dynamic components (S404).

The NHC calculator 123 extracts a Norm Hit Crossing Count (NHC) based on the Norm of the vector extracted from the Norm calculator 122 (S405). At the same time, the NEC calculation unit 124 extracts a Norm Extremum Count (hereinafter referred to as NEC) indicating how much Norm vibrates along the time axis in the concept of the horizontal axis, that is, the time axis in Norm's frame, between NEC and Norm. The Euclidean distance is calculated (S406). The NHC and Euclidean distance extraction methods presented in steps S405 and S406 will be described in detail with reference to FIGS. 5 and 6.

Four types of feature vectors finally extracted through the above-described steps, specifically, Norm, Static Norm, NHC, and Euclidean distance, serve as criteria for determining whether a final fall occurs.

Next, the Norm and Static Norm extracted from the Norm calculator 122 and the NHC values extracted from the NHC calculator 123 are compared with the threshold values th1, th2, and th3 obtained through an iterative experiment, and the NEC calculator 124 In the Euclidean distance signal extracted in step 2), the distance value D at the place where Norm of the maximum peak value is generated is compared with the minimum value th4 in the section in the surrounding frame (S407). As a result of the comparison, when it is determined that the values of the three or more feature vectors are true (S408), this is recognized as a similar fall, and the first emergency call is transmitted to the emergency call gateway 200, and at the same time, the static section detection unit 127 is provided. Give an operation command (S409).

And the static section detection unit 127 receives the operation command from the general section detection unit 126, during the static section (time range of 1 minute to 5 minutes after the first emergency call is transmitted), Norm calculation unit 122 ) Extracts the Norm of the vector, that is, the magnitude component of the entire 3-axis acceleration, and the extracted Norm through the low-pass filter to extract the static Norm signal having reduced dynamic motion components (S404).

Next, the static section detecting unit 127 compares the secondly extracted Norm and Static Norm with experimentally determined thresholds th5 and th6 (S410), and as a result of the comparison, both Norm and Static Norm values have a smaller value than the threshold. If (S411), the user is determined to be almost inoperable and finally recognized as a fall and transmits the second emergency call to the emergency call gateway 200 (S412).

An NHC calculation method for configuring a fall detection method according to an embodiment of the present invention will be described with reference to FIG. 5. 5 is a flowchart illustrating a NHC calculation method of the fall detection method according to an embodiment of the present invention.

First, the maximum range of the amplitude of the longitudinal axis, that is, Norm, is found in a given time interval frame, and the time interval is subdivided from the lowest to the highest interval so that the maximum range can be scanned at a given resolution (S501).

Next, a bias shift from the first interval point (S502) to obtain the biased zero-crossing points (S503), and repeat this to the maximum value interval to store the count for each frame (S504). When one frame is repeated, a distribution of large and small values according to the interval of the magnitude component of Norm can be obtained from the vertical axis, that is, the amplitude point of view.

Finally, the calculated NEC value is converted into a sampling frequency such as Norm and signaled so as to be compared with the Norm value (S505).

An NEC calculation method for configuring a fall detection method according to an embodiment of the present invention will be described with reference to FIG. 6. 6 is a flowchart illustrating a NEC calculation method of the fall detection method according to an embodiment of the present invention.

First, a given frame is divided into subframes along the time axis (S601), while buffering the subframe section (S602), local maximum and minimum values of each subframe section are extracted (S603). The count value of the corresponding subframe is stored (S604).

Next, the count value of the stored subframe is signaled (S605), the Euclidean distance between NEC and Norm is calculated so that the data is a valid value (S606), and the calculated distance value is signaled ( S607).

It will be understood by those skilled in the art that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. It is therefore to be understood that the above-described embodiments are illustrative in all aspects and not restrictive. It is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.

100: wrist worn fall detection device
110: acceleration sensor 120: judgment module
130: wireless communication module 140: location information extraction unit
150: emergency call button 160: wear detection sensor
170: Battery

Claims (10)

Collecting three-axis acceleration and tilt information;
Calculating Norm, which is a magnitude component of all three-axis accelerations, from the collected three-axis acceleration information, and extracting a static Norm having a reduced dynamic component from the Norm;
Calculating a norm hit crossing count (NHC), which means a distribution according to a frame section of the size component of the norm, from the norm;
Calculating, from the Norm, a Norm Extremum Count (NEC) representing a degree of vibration along the time axis of Norm, and calculating a Euclidean distance between the NEC and the Norm; And
Comparing the Norm, the Static Norm, the NEC, and the Euclidean distance with a predetermined threshold value (TH1, TH2, TH3, TH4) to determine whether a fall
Fall detection method comprising a. The method of claim 1, wherein the determining of the fall comprises:
As a result of comparing each of the values of the four feature vectors (Norm, Static Norm, NEC, Euclidean distance) with the threshold value (Norm> TH1, Static Norm> TH2, NHC> TH3, Euclidean distance <TH4) Recognizing that a similar fall has occurred if the value of the three or more feature vectors is determined to be true;
Thereafter, re-extracting Norm and Static Norm using the 3-axis acceleration and slope information collected for a predetermined time; And
Comparing the re-extracted Norm and the Static Norm with a predetermined threshold value (TH5, TH6) to finally determine whether to fall
Fall detection method. The method of claim 2, wherein the finally determining whether the fall,
As a result of comparing the reextracted Norm and the Static Norm with the TH5 and the TH6, respectively, when the values of the Norm and the Static Norm are both smaller than those of the TH5 and the TH6, the fall is finally recognized. Comprising the step of
Fall detection method. The method of claim 3,
Transmitting the first emergency call as a result of recognizing that the similar fall has occurred;
Transmitting a second emergency call as a result of recognizing that the fall has occurred;
Fall detection method further comprising.
The method of claim 1, wherein calculating the NHC,
(a) finding a maximum range of amplitudes of the Norm within a given time interval frame and subdividing the maximum time interval from the lowest to the highest interval;
(b) extracting biased zero-crossing points by biasing from the first interval point in the subdivided time interval;
And repeating step (b) until a subdivided time section corresponding to the highest section and storing a count for each frame.
Fall detection method. The method of claim 1, wherein calculating the NEC,
Dividing a given time interval frame into subframes along the time axis;
Extracting both a local maximum value and a minimum value of each subframe period while buffering the subframe period, and storing a count value of the corresponding subframe;
Fall detection method. In the fall detection device for determining whether the fall, and if it is determined that the fall has occurred, the fall occurrence information is transmitted to the emergency call gateway through wireless communication,
An acceleration sensor for sensing 3-axis acceleration information and tilt information to generate sensing information, and a determination module installed with an application program for determining whether to fall based on the sensing information.
Calculating Norm, which is a magnitude component of all three-axis accelerations, from the collected three-axis acceleration information, and extracting a static Norm having a reduced dynamic component from the Norm;
Calculating a norm hit crossing count (NHC), which means a distribution according to a frame section of the size component of the norm, from the norm;
Calculating, from the Norm, a Norm Extremum Count (NEC) representing a degree of vibration along the time axis of Norm, and calculating a Euclidean distance between the NEC and the Norm; And
And comparing the Norm, the Static Norm, the NEC, and the Euclidean distance with a predetermined threshold value (TH1, TH2, TH3, TH4) to determine whether to fall.
Fall detection device. The method of claim 7, wherein the determining whether the fall,
As a result of comparing each of the values of the four feature vectors (Norm, Static Norm, NEC, Euclidean distance) with the threshold value (Norm> TH1, Static Norm> TH2, NHC> TH3, Euclidean distance <TH4) Recognizing that a similar fall has occurred if the value of the three or more feature vectors is determined to be true;
Thereafter, re-extracting Norm and Static Norm using the 3-axis acceleration and slope information collected for a predetermined time; And
Comparing the re-extracted Norm and the Static Norm with a predetermined threshold value (TH5, TH6) to finally determine whether to fall
Fall detection device. The method of claim 7, wherein
Further comprising a location information extraction unit for collecting the current location information of the user,
The current location information of the user is transmitted to the emergency call gateway together with the fall occurrence information
Fall detection device. The method of claim 7, wherein
The wear detection sensor generates sensing information on whether the user is wearing the fall detection device and transmits it to the emergency call gateway in real time.
Fall detection device further comprising.

Images