CN114847930A - Fall detection device and fall detection method - Google Patents

Abstract

A fall detection device capable of reducing false detection includes: a triaxial acceleration sensor for collecting triaxial acceleration data; and a detection unit for detecting whether a detected object is impacted. The fall detection device further includes a false detection preventing unit that continuously monitors the Y-axis acceleration value in the direction perpendicular to the ground, and for the detected impact, if the Y-axis acceleration value occurs within the removal determination period If it is less than the first threshold value, the false detection preventing unit determines that the fall detection device is removed from the detected object, and the removal determination period is before the impact occurrence time and is the same as the There is a specified time interval between the occurrence time of the shock.

Description

Fall detection device and fall detection method

technical field

The present invention relates to a fall detection device and a fall detection method, in particular to a fall detection device and a fall detection method that reduce false detection.

Background technique

For the elderly, people with limited mobility, etc., there is a high possibility of illness or danger due to a sudden fall. In order to enable caregivers to detect and treat the above-mentioned people in time when they fall, it is known to use a fall detection device. The existing fall detection device is generally worn on the user's body, collects acceleration data of the human body, detects the impact in the collected acceleration data, and determines whether a fall has occurred through the impact. The above impact occurs when the user touches the ground when a fall occurs.

However, the above-mentioned shock may also be caused by other things such as the user's vigorous movement, a collision that occurs, and the like. Therefore, the occurrence of a fall cannot be identified with certainty only by the detection of the shock in the acceleration data. In the prior art, the resultant acceleration and air pressure data obtained from the three-axis acceleration data are simultaneously used to identify a fall. However, as shown in FIGS. 5A to 5D , when the fall detection device is removed, the resultant acceleration map and air pressure The graph is similar to the curve when a fall occurs, that is to say, based on the combined acceleration data and the air pressure data, it is possible to mistakenly recognize the fall detection device as a fall, resulting in a misjudgment of the fall detection. .

In addition, some fall detection devices may verify the occurrence of a fall based on a change in orientation, for example, to detect whether an impact is accompanied by a change in the orientation of the fall detection device (eg, a change from a standing posture to a lying posture). However, the fall detection device is usually worn vertically on the neck or torso of the user only when in use, and is taken off and placed on a table or the like when not in use. Therefore, the change of detection orientation cannot completely exclude the situation that the fall detection device is removed.

Since the fall detection device is often removed, there are many cases of false detection caused by this, and the reliability of the fall detection device is reduced, which brings an extra burden to the caregiver. It is therefore desirable to have a fall detection device that can recognize when itself is removed.

SUMMARY OF THE INVENTION

In order to solve the above technical problems, the present invention provides a fall detection device and a fall detection method that reduce false detection.

A technical solution relates to a fall detection device, comprising: a three-axis acceleration sensor that collects three-axis acceleration data; and a detection unit that detects whether an object to be detected is impacted. The fall detection device is characterized in that it also has a direction perpendicular to the ground. The false detection prevention unit that continuously monitors the Y-axis acceleration value on the upper side, for the impact that has been detected, if the Y-axis acceleration value is smaller than the first threshold value during the removal determination period, the false detection is performed. The prevention unit determines that the fall detection device has been removed from the object to be detected, and the removal determination period is before the impact occurrence time and has a predetermined time interval from the impact occurrence time.

Another technical solution relates to a fall detection method, comprising: the step of collecting triaxial acceleration data, and the step of detecting whether the detected object is impacted, characterized in that it further comprises a false detection prevention step, in the false detection prevention step , the Y-axis acceleration value in the direction perpendicular to the ground is continuously monitored, and for the impact that has been detected, if the Y-axis acceleration value is smaller than the first threshold during the removal determination period, it is determined that In order for the fall detection device to be detached from the object to be detected, the detachment determination period is before the impact occurrence time and has a predetermined time interval from the impact occurrence time.

technical effect

The fall detection device and the fall detection method of the present invention can prevent erroneous detection caused by removal of the fall detection device, improve detection accuracy, and reduce the burden on caregivers.

Description of drawings

FIG. 1 is a schematic diagram of a fall detection device according to the present embodiment.

FIG. 2 is a flowchart showing a fall detection method according to the present embodiment.

FIG. 3 is an explanatory diagram for determining whether or not the fall detection device is removed based on the Y-axis acceleration value in the present embodiment.

4A and 4B respectively show Y-axis acceleration diagrams when a fall occurs and when the fall detection device is removed.

FIGS. 5A to 5D are comparison diagrams between the time of falling and the time when the fall detection device is removed, in the case where fall detection is performed based on the triaxial acceleration value and the air pressure value in the comparative example.

Detailed ways

As has been described above, an impact is generated when the user touches the ground in the event of a fall, but the impact may also be generated by the user's vigorous movement, collisions that occur, and other things. In addition, when a user removes a fall detection device (hereinafter, for the sake of simplicity of description, it may be referred to as a "device") from himself and places it on a support such as a table top, an impact is also generated. If shocks caused by various causes can be distinguished, and shocks caused by causes other than falls can be excluded, the reliability of fall detection can be improved.

Figures 4A and 4B show Y-axis acceleration graphs when a fall occurs and when the device is removed, respectively. When a fall occurs as shown in FIG. 4A , the value of the Y-axis acceleration is relatively stable before the shock representing the action of falling. However, when the device shown in Figure 4B is removed, there is a large change in the Y-axis acceleration due to the pull-up of the device before the shock representing the action of placing the device on the support, because usually The device is only affected by gravity, so the Y-axis acceleration is about 1 gravitational acceleration (1g). However, when the device is pulled up, the Y-axis acceleration is reduced to 0.15g or less due to the upward traction force.

In this embodiment, the variation law of the Y-axis acceleration in the two cases can be used to identify whether the shock is caused by the device being removed, so that the false detection by the fall detection device can be reduced.

Hereinafter, a fall detection device and a fall detection method according to the embodiment will be described with reference to the drawings. In the following embodiment, the schematic configuration of the fall detection device is described as an example of the fall detection device, but the embodiment is not limited to this.

FIG. 1 is a schematic diagram of a fall detection device according to the present embodiment. The fall detection device 100 mainly includes an acceleration sensor 101 , a processing circuit 102 , a communication module 103 and a speaker 104 .

In addition, the fall detection device 100 may be provided with a fixing part externally, such as a vertical clip or a hanging rope. By using the fixing part, the fall detection device can be fixed to the detected object in a manner perpendicular to the ground ( such as the human body).

The acceleration sensor 101 is a sensor that measures X-axis acceleration, Y-axis acceleration, and Z-axis acceleration, which are three-axis accelerations. In the following description, the direction perpendicular to the ground is referred to as the Y direction, the X direction in any direction in a plane parallel to the ground, and the direction perpendicular to the X direction and the Y direction as the Z direction. Acceleration, Y-axis acceleration, and Z-axis acceleration represent accelerations in the X, Y, and Z directions, respectively. In this embodiment, the Y-axis direction is assumed to be the gravitational direction, and the Y-axis acceleration is calibrated using gravitational acceleration (g), which is represented by 1g=512 in each drawing. However, this is just an example, and the X-axis direction or the Z-axis direction may be the direction of gravity.

The processing circuit 102 is an electronic device (for example, a processor) that controls various processes in the fall detection apparatus 100, and in this embodiment, it can execute an impact detection function and a false detection prevention function, and an impact detection function and a false detection prevention function For example, a program executable by a computer is recorded in a memory circuit built in the processing circuit 102 , and the processing circuit 102 reads out and executes each program to realize the function corresponding to each read program.

In the present embodiment, the shock detection function and the erroneous detection prevention function realized by the processing circuit 102 will be described as the shock detection unit 1021 and the erroneous detection prevention unit 1022 .

The shock detection unit 1021 receives the triaxial acceleration data from the acceleration sensor, and generates the resultant acceleration as shown in the upper graph in FIG. 3 , which represents the resultant acceleration of the X axis, the Y axis, and the Z axis, based on the triaxial acceleration data. Then, in the resultant acceleration data, it is detected whether there is a value exceeding the pre-defined shock threshold (second threshold), and if there is a value greater than the shock threshold, it is detected that a shock has occurred, and the information related to the shock is sent to the The erroneous detection prevention unit 1022 transmits and continues detection if there is no value greater than the shock threshold. In this embodiment, the shock threshold is set to 3.5 g, but the shock threshold may be any value from 1.84 g to 5.9 g, where g is the acceleration of gravity.

The false detection prevention unit 1022 continuously monitors the Y-axis acceleration value in the direction perpendicular to the ground (Y direction), and detects whether there is impact detection information from the impact detection unit 1021, In the case of shock detection information, for the detected shock, it is determined whether there is a value less than a predetermined removal threshold (first threshold) during the removal determination period, and if there is a value smaller than the removal threshold, it is determined. For the fall detection device to be removed from the detected object. The removal determination period is before the shock occurrence time, and there is a predetermined time interval with the shock occurrence time, and the predetermined time interval may be a fixed time interval or a time interval corresponding to a fixed sample size. In this embodiment, the removal threshold is set to 0.05g, but the removal threshold may be any value from -0.2g to +0.15g, where g is the acceleration of gravity.

The communication module 103 has functions such as network data communication and telephone communication, and can exchange data with a remote control center. When a fall occurs, the communication module 103 sends a report of the situation to the remote control center.

The speaker 104 may sound an alarm when a fall of the user is detected.

Next, a flowchart of the fall detection method according to the present embodiment will be described with reference to FIG. 2 .

After the fall detection device is started, firstly, three-axis acceleration data is collected in step S101. The collected three-axis acceleration data are used for the generation of the resultant accelerogram and the Y-axis accelerogram. The generation of the resultant acceleration map and the Y-axis acceleration map only needs to be completed before being used, and may be performed in step S101 or in step S102 , so the illustration is omitted in FIG. 2 .

Next, in step S102, using the generated resultant acceleration map, it is determined whether a shock has occurred based on a predefined shock threshold, and if it is determined that no shock has occurred, the process returns to step S101 to continue the collection of triaxial acceleration data. When it is determined that an impact has occurred, the process proceeds to step S103.

Next, in step S103, it is determined whether the fall detection device has been removed from the object to be detected, and if it is determined that it has been removed, the process returns to step S101, and the collection of triaxial acceleration data is continued. When it is determined that it has not been removed, the process proceeds to step S104.

Step S102 and step S103 will be described later with reference to FIG. 3 .

Next, in step S104 , the fall judgment is performed based on other indicators, such as the air pressure value, the orientation, and whether it is in a stationary state, and the like. The other indicators may be one or multiple, and when multiple other indicators are used, fall judgment is performed one by one according to each of the multiple other indicators. In S105, it is judged whether the above-mentioned other indicators all meet the judgment standard of falling. If any of the above-mentioned indicators does not meet the judgment standard of falling, return to step S101. When all the indicators meet the judgment standard of falling, proceed to step 106, and judge at this time. for a fall and sound an alarm.

The steps of S101, S104, S105, and S106 described above are the same as those in the prior art, so detailed descriptions are omitted. Only the different steps are detailed below.

FIG. 3 is an explanatory diagram for determining whether or not the fall detection device is removed based on the Y-axis acceleration value in the present embodiment. FIG. 3 corresponds to steps S102 to S103 , and the following description will be made with reference to FIG. 3 . The graph of FIG. 3 is obtained by sampling data for 6 seconds at a sampling frequency of 50 Hz, for example.

As shown in FIG. 3 , after the resultant acceleration data is generated based on the triaxial acceleration data from the acceleration sensor 101 , the impact detection unit 1021 detects whether the resultant acceleration data exceeds a predetermined impact threshold (refer to the upper diagram of FIG. 3 ). The horizontal dotted line in this embodiment is the value of 3.5g). After detecting the resultant acceleration value exceeding the shock threshold, it is determined that a shock has occurred at that time, and this time is taken as the shock occurrence time. In FIG. 3 , the time of occurrence of the shock is represented by the character S.

After that, the erroneous detection preventing unit 103 judges, in the Y-axis acceleration data, whether or not there is less than the detachment threshold value in the detachment determination period T2, which is the detachment judgment period (refer to the horizontal dotted line in the lower diagram of FIG. 3 , this embodiment 0.05g in the mode). The removal determination period T2 is before the shock occurrence time S, and is separated from the shock occurrence time S by a predetermined interval time T1 determined in advance, the predetermined interval time T1 being to exclude a period in which data before the shock occurrence is unstable Whereas, the predetermined interval time T1 may be a fixed time interval, or may be a time interval corresponding to a fixed sample size. The above-mentioned removal determination period T2 and predetermined interval time T1 are set according to actual needs. As an example, in this embodiment, the predetermined interval time T1 is the interval time corresponding to 15 samples, and the removal determination period T2 is: The period corresponding to the 15th to 70th sample sizes before the shock occurrence time S, in other words, is the period of 0.3 to 1.7 seconds before the shock occurrence time S. FIG.

In the present embodiment, the resultant acceleration is used to determine the occurrence of a shock, and the Y-axis acceleration is used to determine whether or not the generated shock is a shock caused by the fall detection device being removed. Accordingly, if the shock is caused by the fall detection device being removed, the shock is excluded and subsequent fall detection determination is not performed, thereby reducing the occurrence of false detection and improving the reliability of the fall detection device.

(Comparative example)

As an example of the comparative example, FIG. 5A to FIG. 5D show a case where fall detection is performed using the combined acceleration value and the air pressure value under the same conditions. 5A shows a resultant acceleration diagram when a fall occurs, FIG. 5B shows a resultant acceleration diagram when the fall detection device is removed, FIG. 5C shows an air pressure diagram when a fall occurs, and FIG. 5D shows an air pressure diagram when the fall detection device is removed. By comparing FIGS. 5A and 5B, and FIGS. 5C and 5D, it can be seen that under the same conditions, the resultant acceleration value and the air pressure value are indistinguishable in the two cases, so that the action of removing the fall detection device cannot be removed from the entire fall. Excluded from the detection, resulting in the occurrence of false detection.

On the other hand, in the present embodiment, the occurrence of the shock is determined using the resultant acceleration, and whether or not the shock that has occurred is caused by the fall detection device being removed is determined using the Y-axis acceleration. When the fall detection device is removed, the shock can be excluded from the fall detection.

According to the experimental results of the researchers, the use of the fall detection device of the present embodiment can prevent about 95% of false alarms when the fall detection device is removed, compared with the case of the comparative example. Thereby, the reliability of the fall detection device is improved.

As mentioned above, although the specific embodiment of this invention was described, this embodiment is shown as an example, Comprising: It does not intend to limit the scope of the invention. Each component in the embodiment can be implemented in other various forms, and various omissions, substitutions, and changes can be made without departing from the gist of the invention. These embodiments and modifications thereof are included in the scope and spirit of the invention, and are included in the invention described in the claims and the scope of equivalents thereof.

Claims (10)

1. A fall detection device, comprising: A triaxial acceleration sensor that collects triaxial acceleration data; and The detection part detects whether the detected object is impacted, It is characterized by further comprising a false detection preventing unit, The false detection prevention unit continuously monitors the Y-axis acceleration value in the direction perpendicular to the ground, and for the impact that has been detected, if the Y-axis acceleration value is smaller than the first threshold during the removal determination period. In this case, it is determined that the fall detection device is detached from the detected object, and the detachment determination period is before the impact occurrence time and has a predetermined time interval from the impact occurrence time. 2. The fall detection device according to claim 1, wherein, The first threshold is any value between -0.2g and +0.15g, where g is the acceleration of gravity. 3. The fall detection device of claim 2, wherein The first threshold is 0.05g. 4. The fall detection device of claim 1, wherein When the resultant acceleration obtained by the triaxial acceleration sensor is greater than the second threshold value, it is determined that the shock has occurred. 5. The fall detection device of claim 4, wherein The second threshold is any value from 1.84g to 5.9g, where g is the acceleration of gravity. 6. The fall detection device of claim 5, wherein The second threshold is 3.5g. 7. The fall detection device of claim 1, wherein Also has a fixed part, The fixing portion fixes the fall detection device to the object to be detected so as to be perpendicular to the ground. 8. The fall detection device of claim 7, wherein The fixing part is a clip or a neck hanging rope. 9. The fall detection device of claim 1, wherein The predetermined time interval is a fixed time interval or a time interval corresponding to a fixed sample size. 10. A fall detection method, comprising: a step of collecting triaxial acceleration data, and a step of detecting whether a detected object is impacted, characterized in that: Also includes false detection prevention steps, In the false detection prevention step, the Y-axis acceleration value in the direction perpendicular to the ground is continuously monitored, and for the detected impact, if the Y-axis acceleration value is less than the first value during the removal determination period In the case of a threshold value, it is determined that the fall detection device has been removed from the detected object, and the removal determination period is before the impact occurrence time and has a predetermined time interval from the impact occurrence time.

Images