CN104000596B - A fall detection method based on mobile terminal - Google Patents

Abstract

The invention discloses a kind of fall detection methods based on mobile terminal, comprise the following steps: detection resultant acceleration SVM1, judge whether resultant acceleration SVM1 has been more than to set weightless threshold value A min；As SVM1 < Amin, compound direction signal Ω 1 is detected, judges whether direction signal Ω 1 is more than setting tilt angle threshold value Ω min；As Ω 1 < Ω min, close angle speed omega is detected, judges whether the close angle speed omega is more than set angle threshold speed ω min；Work as ω > ω min, detects resultant acceleration SVM2；In setting detection time t1, as resultant acceleration SVM2 > setting impact acceleration threshold value A max, compound direction signal Ω 2 is detected, judges whether direction signal Ω 2 is more than the human body tumble gradient threshold value Ω max set；As Ω 2 < Ω max, illustrate that people has fallen；The signal processing that will be deemed as human body tumble forms voice prompting signal, image display signal.Characteristic acceleration during the present invention falls with reference to human body, human body heeling condition, angular speed etc. obtain the parameter needed using sensor built-in in mobile phone, analyze and determine whether human body falls.

Description

A fall detection method based on mobile terminal

technical field

The invention relates to the identification of human body fall conditions, in particular to a fall detection method based on a mobile terminal.

Background technique

With the gradual increase of my country's population aging, there are more and more empty-nesters in society. And the old people's physique is weaker, often can accidentally fall, and fall itself has caused great harm to the old people, and not timely rescue can bring fatal danger to the old people more. Therefore, various methods of effectively identifying human body falls and reporting to the police in time are of great significance to society. These include: based on video analysis technology, real-time monitoring and analysis of the human body to determine whether the human body has fallen; based on audio signal analysis technology, the fall status is judged by analyzing the frequency when the person hits the ground and other judgment methods.

As smartphones have become one of the people's portable devices, it has become a more convenient and effective method to identify human falls based on smartphones, which can well reduce the injuries caused by the failure of timely assistance for the elderly after falls.

Contents of the invention

The purpose of the present invention is to provide a fall detection method based on a mobile terminal. People's daily exercise behavior mainly includes walking, standing up, sitting down, running and so on. There is no weightless feature when walking and standing up. The combined acceleration of the acceleration sensor when sitting down is much smaller than the acceleration when hitting. The body tilt angle changes little and has no angular velocity when running. Therefore, the above-mentioned characteristics can well distinguish human falls from other daily motor behaviors. The present invention detects whether the human body has fallen by using various sensors built in the intelligent mobile terminal, and confirms whether the judgment is correct by touching the display of the mobile terminal. Ask for help; it can quickly and effectively avoid bodily injury caused by failure to receive timely assistance after a fall.

In order to achieve the above object, the present invention is achieved through the following technical solutions:

A fall detection method based on a mobile terminal is characterized in that the method comprises the following steps:

Step S1, detecting the resultant acceleration SVM1, and judging whether the resultant acceleration SVM1 exceeds the set weightlessness threshold Amin; when SVM1<Amin, execute step S2.

Step S2 , detecting the synthesized direction signal Ω1 , judging whether the direction signal Ω1 exceeds the set threshold Ωmin of the inclination angle between the human body and the ground, and when Ω1<Ωmin, go to step 3 .

Step S3, detecting the combined angular velocity ω, and judging whether the combined angular velocity ω exceeds the set angular velocity threshold ωmin; when ω>ωmin, execute step S4.

Step S4 , detecting the resultant acceleration SVM2 , and within the set detection time t1 , when the resultant acceleration SVM2 exceeds the set impact acceleration threshold Amax, step 5 is executed.

Step S5, detecting the composite direction signal Ω2, and judging whether the direction signal Ω2 exceeds the set human fall inclination threshold Ωmax; when Ω2<Ωmax, it means that the person has fallen, and step S6 is executed.

Step S6, processing the signal judged to be a human body fall to form a voice prompt signal and an image display signal respectively.

The above-mentioned fall detection method based on the mobile terminal is characterized in that the above-mentioned step S1 is specifically performed as follows: the three-axis acceleration sensor detects the combined acceleration SVM1, and transmits the detected result to the CPU, and the above-mentioned CPU judges whether the combined acceleration SVM1 exceeds The set weightless threshold Amin; when the combined acceleration SVM1 does not exceed the weightless threshold Amin, step S2 is executed; otherwise, the above-mentioned three-axis acceleration sensor detects the combined acceleration SVM1 again.

The above-mentioned fall detection method based on a mobile terminal is characterized in that in the above-mentioned step S2, the above-mentioned CPU starts the above-mentioned three-axis acceleration sensor and the magnetic sensor to obtain detection signals respectively and transmits them to the above-mentioned CPU, and the CPU uses the above-mentioned two sensors. The detection signal synthesizes the direction signal Ω1, and judges whether the direction signal Ω1 exceeds the set threshold Ωmin of the inclination angle between the human body and the ground. When Ω1<Ωmin, perform step 3; when Ω1≥Ωmin, jump to step S1.

The above-mentioned fall detection method based on the mobile terminal is characterized in that in the above-mentioned step S3, the above-mentioned CPU starts the gyroscope sensor to detect the combined angular velocity ω, and transmits the combined angular velocity ω to the above-mentioned CPU, and the CPU judges whether the combined angular velocity ω exceeds The set angular velocity threshold ωmin; when ω>ωmin, execute step S4; otherwise, jump to step S1.

The above-mentioned fall detection method based on the mobile terminal is characterized in that in the above-mentioned step S4, the above-mentioned CPU starts the above-mentioned three-axis acceleration sensor to detect the combined acceleration SVM2, and within the detection time t1 set by the CPU, when the combined acceleration SVM2>set When the impact acceleration threshold Amax is determined, execute step 5; otherwise, jump to step S1.

The above-mentioned fall detection method based on the mobile terminal is characterized in that in the above-mentioned step S5, the above-mentioned CPU restarts the above-mentioned three-axis acceleration sensor and the magnetic sensor to detect respectively, and the CPU processes the above-mentioned two detection signals to synthesize the direction signal Ω2, and Judging whether the direction signal Ω2 exceeds the set human body fall inclination threshold Ωmax; when Ω2<Ωmax, it means that the person has fallen, and executes step S6; otherwise, skips to step S1.

The above-mentioned fall detection method based on the mobile terminal is characterized in that in the above-mentioned step S6, the above-mentioned CPU processes the signal that is judged to be a human body fall to form a voice prompt signal and an image display signal, and transmits them to the speaker and the touch display of the mobile terminal respectively.

The above-mentioned fall detection method based on the mobile terminal is characterized in that, when the mobile terminal is dormant, the above-mentioned CPU starts the distance sensor to detect whether there is an obstacle in front of the mobile terminal; The mobile terminal transmits the signal to the CPU, so that the CPU starts the above-mentioned three-axis acceleration sensor to perform the above-mentioned step S1; when the distance sensor does not detect an obstacle signal, it is judged that the mobile terminal is separated from the human body, and there is no need to start The three-axis acceleration sensor.

The above mobile terminal-based fall detection method is characterized in that, when the mobile terminal is in the start-up or running state, the above-mentioned three-axis acceleration sensor automatically starts to detect the combined acceleration SVM1.

The above-mentioned fall detection method based on the mobile terminal is characterized in that the above-mentioned step S6 also includes the following steps:

Step S6.1, within the set response time t2, when the above-mentioned mobile terminal touch display does not transmit to the above-mentioned CPU to confirm that the information is a false alarm, the CPU starts the GPS module for human body positioning, and the CPU receives the above-mentioned GPS module. Locate the signal to and ask for help through text messages or phone calls;

Step S6.2: within the set response time t2, when the mobile terminal touches the display and transmits to the CPU to confirm that the information is a false positive, the operation ends.

Compared with the prior art, the present invention has the following advantages:

1. By analyzing the characteristics of the state of motion changes in the process of human falls, and find out the difference between these characteristics and the characteristics of daily life movement. The features mainly referred to in the present invention are acceleration, human body tilt state, angular velocity and so on.

2. According to the previous analysis, use the built-in sensors in the mobile phone to obtain the required parameters, and analyze whether it matches the situation of the fall.

3. Finally, judge whether you have fallen. If you fall, you will use GPS to locate the information, and send the information to the person designated in advance through your mobile phone, waiting for help.

Description of drawings

FIG. 1 is an overall flow chart of a mobile terminal-based fall detection method of the present invention.

FIG. 2 is a schematic diagram of the system structure of a mobile terminal-based fall detection method according to the present invention.

Detailed ways

The present invention will be further elaborated below by describing a preferred specific embodiment in detail in conjunction with the accompanying drawings.

As shown in Figure 2, the fall detection system based on mobile terminal comprises: CPU 10, and the three-axis acceleration sensor 20 that is connected with it, gyroscope sensor 30, distance sensor 40, magnetic force sensor 50, GPS module 60, loudspeaker 70 and mobile terminal Touch the display 80 . The CPU 10 performs two-way communication with the above modules respectively. The detection system is built into the smart mobile terminal.

In real life, people generally place their mobile phones in the pockets on both sides of their trousers or coat, close to their waists, which can well reflect the movement state of the human body. The default position of the mobile phone in the present invention is placed at this position.

There are four common ways for the human body to fall, which are forward fall, backward fall, left fall, and right fall. In these types of falls, there will be several important features: weightlessness, the human body will go from standing upright to tilting until it is nearly parallel to the ground, and hit the ground. According to the above several features, the present invention realizes the detection of human body falls through the above-mentioned fall detection system arranged in the intelligent mobile terminal.

As shown in Figure 1, a kind of fall detection method based on mobile terminal, this method comprises the following steps:

Step S1, the three-axis acceleration sensor 20 detects the combined acceleration SVM1, and transmits the detected result to the CPU 10, and the above-mentioned CPU 10 judges whether the combined acceleration SVM1 exceeds the set weightlessness threshold Amin; when the combined acceleration SVM1 does not exceed the weightlessness When the threshold Amin is reached, step S2 is executed; otherwise, the above-mentioned three-axis acceleration sensor 20 detects the resultant acceleration SVM1 again.

In this embodiment, the weight loss threshold Amin=0.4g (g is the acceleration of gravity) is set.

Under normal conditions of the mobile terminal, the resultant acceleration detected by the triaxial acceleration sensor 20 is close to the gravitational acceleration g; in the process of falling, the human body will experience weightlessness. Then, during the falling process, the resultant acceleration detected by the triaxial acceleration sensor 20 is obviously smaller than the gravitational acceleration g. Among them, the algorithm of combined acceleration is as follows:

are the accelerations of the three-axis acceleration sensor 20 on the x-axis, y-axis and z-axis respectively.

Step S2, the CPU 10 starts the above-mentioned three-axis acceleration sensor 20 and the magnetic sensor 50 to obtain detection signals respectively and transmits them to the above-mentioned CPU 10, and the CPU 10 synthesizes the detection signals obtained by the above-mentioned two sensors into a direction signal Ω1, and judges the direction Whether the signal Ω1 exceeds the set threshold Ωmin of the inclination angle between the human body and the ground, when Ω1<Ωmin, go to step 3; when Ω1≥Ωmin, jump to step S1.

When the human body is standing, the angle between the human body and the ground is about 90°; when it falls, the angle between the human body and the ground continues to decrease, and when it falls to the ground, the angle is about 0°. In the present invention, the detection signals respectively generated by the triaxial acceleration sensor 20 and the magnetic sensor 50 are processed by the CPU 10 to form a direction signal Ω1, that is, the angle Ω1 between the human body and the ground can be obtained.

In this embodiment, the value of Ωmin is .

Step S3, the CPU 10 starts the gyro sensor 30 to detect the combined angular velocity ω, and transmits the combined angular velocity ω to the CPU 10, and the CPU 10 judges whether the combined angular velocity ω exceeds the set angular velocity threshold ωmin; when ω>ωmin, execute the step S4; otherwise, go to step S1.

During the fall process of the human body, a larger combined angular velocity ω will be generated. The combined angular velocity ω is the combined velocity of the angular velocities generated by the gyro sensor 30 in the three directions of the x-axis, y-axis and z-axis. The algorithm is as follows:

are the angular velocities generated by the gyro sensor 30 in the three directions of x-axis, y-axis and z-axis, respectively. In this embodiment, ωmin is set to be π/6 rad/s.

Step S4, the CPU 10 starts the three-axis acceleration sensor 20 to detect the combined acceleration SVM2, and within the detection time t1 set by the CPU 10, when the combined acceleration SVM2 exceeds the set impact acceleration threshold Amax, execute step 5; otherwise, skip Go to step S1.

During the collision between the human body and the ground, a large combined acceleration SVM2 will be generated. The test found that the maximum combined acceleration SVM2 generated during this process is 12.9 times the gravitational acceleration g (12.9g), and the minimum combined acceleration SVM2 is 4.9g, with an average of 6.1g. Therefore, in the present invention, the CPU 10 restarts the three-axis acceleration sensor 20 to detect the combined acceleration SVM2. In this embodiment, Amax=4.9g. When SVM2>4.9g within the set detection time t1, it is determined that the human body is in contact with the ground. A collision has occurred; otherwise, return to step S1 and perform detection again.

Step S5, the above-mentioned CPU 10 restarts the above-mentioned three-axis acceleration sensor 20 and the magnetic sensor 50 to detect respectively, the CPU 10 processes the above-mentioned two detection signals to synthesize the direction signal Ω2, and judges whether the direction signal Ω2 exceeds the set value Threshold Ωmax for the inclination of the human body falling; when Ω2<Ωmax, it means that the person has fallen, go to step S6; otherwise, go to step S1.

In this embodiment, the value of Ωmax is .

In step S6, the above-mentioned CPU 10 processes the signal judged to be a human fall to form a voice prompt signal and an image display signal, and transmits them to the speaker 70 and the touch display 80 of the mobile terminal respectively. This step includes the following steps:

Step S6.1, within the set response time t2, when the mobile terminal touch display 80 does not transmit to the CPU 10 to confirm that the information is a false alarm, the CPU 10 starts the GPS module 60 for human body positioning, and the CPU 10 receives the GPS module 60 Locate the signal to and ask for help through text messages or phone calls;

Step S6.2: within the set response time t2, when the mobile terminal touches the display 80 to transmit to the CPU 10 and confirms that the information is a false positive, the operation ends.

The above step S6 can reduce the false alarm rate, thereby avoiding subsequent actions caused by false alarms.

In a mobile terminal-based fall detection method provided by the present invention, before performing step S1, when the mobile terminal is in sleep mode, the CPU 10 starts the distance sensor 40 to detect whether there is an obstacle in front of the mobile terminal; when the distance sensor 40 detects an obstacle signal When it is determined that the human body carries the mobile terminal, the signal is transmitted to the CPU 10, so that the CPU 10 starts the triaxial acceleration sensor 20 to perform the above-mentioned step S1; when the distance sensor 40 does not detect an obstacle signal, it is judged that the mobile terminal The terminal is separated from the human body, and the triaxial acceleration sensor 20 does not need to be activated.

Before step S1 is executed, when the mobile terminal is in the start-up or running state, the above-mentioned three-axis acceleration sensor 20 performs the above-mentioned step S1, and automatically starts to detect the resultant acceleration SVM1.

Through the above settings, the method can reduce the consumption of the battery of the mobile terminal.

Although the content of the present invention has been described in detail through the above preferred embodiments, it should be understood that the above description should not be considered as limiting the present invention. Various modifications and alterations to the present invention will become apparent to those skilled in the art upon reading the above disclosure. Therefore, the protection scope of the present invention should be defined by the appended claims.

Claims (9)

1. a kind of fall detection method based on mobile terminal, which is characterized in that this method comprises the following steps:

Step S1 detects resultant acceleration SVM1, judges whether resultant acceleration SVM1 has been more than the weightless threshold value A min set；When When SVM1 < Amin, step S2 is executed；

Step S2 detects compound direction signal Ω 1, judges whether direction signal Ω 1 is more than the human body of setting and inclining for ground Rake angle threshold value Ω min executes step S3 as Ω 1 < Ω min；

Step S3, detect close angle speed omega, judge the close angle speed omega whether be more than setting angular speed threshold value ω min；Work as ω > When ω min, step S4 is executed；

Step S4 detects resultant acceleration SVM2, in the detection time t1 of setting, when resultant acceleration SVM2 is more than hitting for setting When hitting acceleration rate threshold Amax, step S5 is executed；

Step S5 detects compound direction signal Ω 2, and judges whether direction signal Ω 2 is more than the human body tumble inclination set Spend threshold value Ω max；As Ω 2 < Ω max, illustrate that people has fallen, executes step S6；

Step S6, the signal processing that will be deemed as human body tumble are respectively formed voice prompting signal, image display signal；

Before executing the step S1, when mobile terminal sleep, it is mobile eventually that CPU (10) starts range sensor (40) detection Whether end front has barrier；When the range sensor (40) detects obstacle signal, it is judged as that human body carries and is somebody's turn to do The signal is then transmitted to the CPU (10) by mobile terminal, so that the CPU (10) starting 3-axis acceleration sensor (20) carries out The step S1；When obstacle signal is not detected in the range sensor (40), judge that mobile terminal is separated with human body, nothing The 3-axis acceleration sensor (20) need to be started.

2. as described in claim 1 based on the fall detection method of mobile terminal, which is characterized in that the step S1 is specific It executes as follows:

3-axis acceleration sensor (20) detects resultant acceleration SVM1, and the result that will test is transmitted to CPU (10), described CPU (10) judges whether resultant acceleration SVM1 has been more than the weightless threshold value A min set；When resultant acceleration SVM1 is less than this When weightless threshold value A min, step S2 is executed；Otherwise the 3-axis acceleration sensor (20) detects resultant acceleration SVM1 again.

3. as claimed in claim 2 based on the fall detection method of mobile terminal, which is characterized in that the step S2 is specific It executes as follows:

CPU (10) starting 3-axis acceleration sensor (20) and magnetometric sensor (50) obtain detection letter respectively Number and be transmitted to the CPU (10), which obtains 3-axis acceleration sensor (20) and magnetometric sensor (50) Signal compound direction signal Ω 1 is detected, and judges whether direction signal Ω 1 is more than the human body of setting and the tilt angle on ground Threshold value Ω min executes step S3 as Ω 1 < Ω min；As Ω 1 >=Ω min, go to step S1.

4. as claimed in claim 3 based on the fall detection method of mobile terminal, which is characterized in that the step S3 is specific It executes as follows:

CPU (10) starting gyro sensor (30) detects close angle speed omega, and the close angle speed omega is transmitted to institute The CPU (10) stated, the CPU (10) judge the close angle speed omega whether be more than setting angular speed threshold value ω min；Work as ω > ω min When, execute step S4；Otherwise go to step S1.

5. as claimed in claim 4 based on the fall detection method of mobile terminal, which is characterized in that the step S4 is specific It executes as follows:

CPU (10) starting 3-axis acceleration sensor (20) detects resultant acceleration SVM2, sets in the CPU (10) In fixed detection time t1, as resultant acceleration SVM2 > setting impact acceleration threshold value A max, step S5 is executed；Otherwise Go to step S1.

6. as claimed in claim 5 based on the fall detection method of mobile terminal, which is characterized in that the step S5 is specific It executes as follows:

The CPU (10) is again started up the 3-axis acceleration sensor (20), magnetometric sensor (50) is examined respectively It surveys, which believes the detection signal processing compound direction that 3-axis acceleration sensor (20), magnetometric sensor (50) obtain Number Ω 2, and judge direction signal Ω 2 whether be more than setting human body tumble gradient threshold value Ω max；As Ω 2 < Ω max, Illustrate that people has fallen, executes step S6；Otherwise go to step S1.

7. as claimed in claim 6 based on the fall detection method of mobile terminal, which is characterized in that the step S6 is specific It executes as follows:

It is aobvious that the signal processing that step S6, the CPU (10) will be deemed as human body tumble is respectively formed voice prompting signal, image Show signal, and is transmitted separately to loudspeaker (70), mobile terminal touch display (80).

8. as described in claim 1 based on the fall detection method of mobile terminal, which is characterized in that executing the step Before S1, when mobile terminal is in starting or operating status, the 3-axis acceleration sensor (20) automatically begins to carry out The step S1.

9. as claimed in claim 7 based on the fall detection method of mobile terminal, which is characterized in that the step S6 is also wrapped Containing following steps:

Step S6.1, in the response time t2 of setting, the mobile terminal touch display (80) is not to the CPU (10) when transmission confirms that the information is wrong report, which starts GPS module (60) and carries out human body positioning, which receives The positioning signal of the GPS module (60) is sought help to and by short message or phone, terminates this operation；

Step S6.2, in the response time t2 of setting, which transmits to the CPU (10) confirms When the information is wrong report, terminate this operation.