CN220212910U - Human body fall detection device based on infrared array - Google Patents

Abstract

The utility model discloses a human body falling detection device based on an infrared array in the field of electronic information communication, which comprises an infrared array sensor, wherein the infrared array sensor is connected with a singlechip, the singlechip is respectively connected with an Internet of things module and a GSM module, and the infrared array sensor, the singlechip, the Internet of things module and the GSM module are all connected with a power supply module; human body temperature data detected by the infrared array sensor are collected by the singlechip and classified, so that falling behaviors are identified and judged. If the falling action is identified, the equipment is connected with the internet of things platform through the internet of things module, can receive alarm information on the mobile phone APP, and can send alarm short messages to the appointed mobile phone through the GSM module, so that dual alarm reminding is realized.

Description

Human body fall detection device based on infrared array

Technical Field

The utility model relates to a human body fall detection device in the field of electronic information communication.

Background

The falling is a serious threat to the health of the old, the accidental falling of the old is accurately detected and rescue is timely carried out, and the risks of death and injury can be reduced to the minimum. In recent years, various techniques have been widely used in the fall detection field in combination with machine learning and deep learning algorithms. Currently common fall detection systems or devices fall broadly into three categories: wearable-based devices, environmental-based sensors, and video-sensor-based fall detection devices. However, in the current fall detection scheme, the wearing behavior of the fall monitoring device is slightly complicated, and part of people do not have the behavior habits of wearing watches, ornaments and the like. Because of the small requirement on the wearable hardware, the problem of electric quantity endurance cannot be effectively improved. Especially in the face of the elderly use field Jing Shi, the relatively poor memory of the elderly is a big problem, and the old forgets to charge and wear the device. Only wearing the system can enable the system to exert the functions of monitoring and alarming, so that the problems of memory and living habits of users become a huge hidden trouble that the system cannot exert the effects of the system. Because the fall monitoring of the visual identification technology must comprise a camera, the worry of the user about privacy disclosure is caused, the whole system has higher privacy disclosure risk due to the characteristics of the internet of things, and the price of the equipment is relatively higher due to the requirements of the video processing and identification system on performance.

Disclosure of Invention

The utility model aims to provide a human body falling detection device based on an infrared array, which realizes the falling detection function of safety, low cost and no using threshold.

In order to achieve the above purpose, the utility model provides a human body falling detection device based on an infrared array, which comprises an infrared array sensor, wherein the infrared array sensor is connected with a singlechip, the singlechip is respectively connected with an Internet of things module and a GSM module, and the infrared array sensor, the singlechip, the Internet of things module and the GSM module are all connected with a power supply module.

Compared with the prior art, the intelligent falling-down detection system has the beneficial effects that the singlechip is used for collecting the human body temperature data detected by the infrared array sensor and classifying the human body temperature data so as to identify and judge falling-down behaviors. If the falling action is identified, the equipment is connected with the internet of things platform through the internet of things module, can receive alarm information on the mobile phone APP, and can send alarm short messages to the appointed mobile phone through the GSM module, so that dual alarm reminding is realized.

As a further improvement of the utility model, the singlechip comprises a chip STM32, the infrared array sensor comprises a chip AMG8833, a pin 2 of the chip STM32 is connected with a pin 93 of the chip STM32, a pin 3 of the chip STM32 is connected with a pin 92 of the chip STM32, a pin 6 of the chip STM32 is grounded, pins 13 and 9 of the chip STM32 are connected with a power supply module, a pin 10 of the chip STM32 is connected with a pin 12 of the chip STM32 through a resistor R4, capacitors C9 and C10, one end of a capacitor C9 is connected with one end of a capacitor C7 in a grounding manner, the other end of the capacitor C7 is connected with a pin 9 of the chip STM32, and the capacitor C7 is connected with a capacitor C8 in parallel.

In this way, three times of data acquisition are carried out every second through the chip AMG8833, the data format of each time is 8 x 8 temperature data array, the AMG8833 transmits the data to the single chip microcomputer chip STM32 in real time through the IIC interface, and the chip STM32 carries out calculation, feature extraction and data classification judgment on the data.

As a further improvement of the utility model, the Internet of things module comprises a chip ESP8266, a pin 1, a pin 3, a pin 4, a pin 29 and a pin 30 of the chip ESP8266 are connected with the power supply module, a pin 2 of the chip ESP8266 is connected with the antenna, a pin 7 of the chip ESP8266 is connected with a pin 11, a pin 28, a pin 50, a pin 75 and a pin 100 of the chip STM32 through a resistor R6, a pin 26 of the chip ESP8266 is connected with a pin 25 of the chip STM32, and a pin 25 of the chip ESP8266 is connected with a pin 26 of the chip STM32.

Therefore, after the STM32 singlechip judges the falling action, alarm information is sent to the ESP8266 through the serial port 1, the ESP8266 transmits data to the Ali cloud server through the home network and the router by the MQTT protocol, the alarm information can be received on the mobile phone APP, the process is accurate and rapid, and a caretaker can be timely reminded of falling to the caretaker.

As a further improvement of the utility model, the GSM module comprises a chip SIM900A, wherein the J1.1 pin of the chip SIM900A is connected with the power supply module, the J1.2 pin of the chip SIM900A is connected with the J1.7 pin of the chip SIM900A and grounded, the J1.5 pin of the chip SIM900A is connected with the 68 pin of the chip STM32, and the J1.6 pin of the chip SIM900A is connected with the 69 pin of the chip STM32.

Therefore, after judging the falling action, the STM32 singlechip transmits alarm information to the SIM900A module through the serial port 2, and the SIM900A receives the alarm information and transmits alarm short messages to the mobile phone numbers set in the program, so that double insurance can be achieved, and caregivers can be reminded in multiple ways.

As a further improvement of the utility model, the power supply module comprises a voltage stabilizing chip 1117A, wherein the pin 1 and the pin 3 of the voltage stabilizing chip 1117A are connected with the pin 3 of an isolating switch power supply U8, the pin 4 of the isolating switch power supply U8 is grounded, the pin 1 and the pin 2 of the isolating switch power supply U8 are respectively connected with the pin 1 and the pin 2 of a socket J1, the pin 1 of the voltage stabilizing chip 1117A is also connected with the pin J1.1 of a chip SIM900A, a capacitor C4 is connected between the pin 1 and the pin 4 of the voltage stabilizing chip 1117A, the capacitor C4 is connected with a capacitor C3 in parallel, the capacitor C3 is connected with a light emitting diode and a resistor R1 which are connected in series, the pin 4 of the voltage stabilizing chip 1117A is respectively connected with the pin 4, the pin 5, the pin 13 and the pin 9 of a chip AMG8833, the pin 4 of the voltage stabilizing chip 1117A is also connected with a capacitor C11, a capacitor/12 capacitor C13 and a capacitor C14 which are mutually connected in parallel, and the pins 824 of the voltage stabilizing chip is connected with pins 1, 66, 29 and 30.

Socket J1 connects 220V alternating current power, converts 220V alternating current into +5v power through isolator power supply U8, and then provides stable 3.3V voltage for each chip through steady voltage chip 1117A, ensures the normal use of equipment.

Drawings

FIG. 1 is a schematic diagram of the basic composition of the present utility model.

Fig. 2 is a schematic diagram of the overall device connection of the present utility model.

Fig. 3 is a flow chart of the fall detection algorithm according to the utility model.

Fig. 4 is a circuit hardware connection diagram of the present utility model.

Fig. 5 is a diagram of array pixels of the present utility model when properly stood by a caretaker.

Fig. 6 is a diagram of array pixels of a caretaker of the present utility model falling.

Description of the embodiments

The utility model is further described below with reference to the accompanying drawings:

the human body falling detection device based on the infrared array shown in the figures 1-4 comprises an infrared array sensor, wherein the infrared array sensor is connected with a singlechip, the singlechip is respectively connected with an Internet of things module and a GSM module, and the infrared array sensor, the singlechip, the Internet of things module and the GSM module are all connected with a power supply module.

The singlechip includes chip STM32, infrared array sensor includes chip AMG8833, the No. 2 foot of chip STM32 links to each other with the 93 foot of chip STM32, the No. 3 foot of chip STM32 links to each other with the 92 foot of chip STM32, the 6 foot ground connection of chip STM32, the 13 foot and the 9 foot of chip STM32 link to each other with power module, the 10 foot of chip STM32 links to each other with the 12 foot of chip STM32 through resistance R4, electric capacity C9 and C10, electric capacity C9 one end links to each other with electric capacity C7 one end ground connection, the electric capacity C7 other end links to each other with the 9 foot of chip STM32, electric capacity C7 connects in parallel has electric capacity C8.

The internet of things module includes chip ESP8266, and No. 1 foot, no. 3 foot, no. 4 foot, no. 29 foot and No. 30 foot of chip ESP8266 link to each other with power module, and No. 2 foot and the antenna of chip ESP8266 link to each other, and No. 7 foot of chip ESP8266 links to each other with 11 feet, no. 28 feet, no. 50 feet, no. 75 feet and No. 100 feet of chip STM32 through resistance R6, and No. 26 feet of chip ESP8266 link to each other with No. 25 feet of chip STM32, and No. 25 feet of chip ESP8266 link to each other with No. 26 feet of chip STM32.

The GSM module comprises a chip SIM900A, wherein a J1.1 pin of the chip SIM900A is connected with the power supply module, a J1.2 pin of the chip SIM900A is connected with a J1.7 pin of the chip SIM900A and grounded, a J1.5 pin of the chip SIM900A is connected with a 68 pin of the chip STM32, and a J1.6 pin of the chip SIM900A is connected with a 69 pin of the chip STM32.

The power supply module comprises a voltage stabilizing chip 1117A, a pin 1 and a pin 3 of the voltage stabilizing chip 1117A are connected with a pin 3 of an isolating switch power supply U8, a pin 4 of the isolating switch power supply U8 is grounded, a pin 1 and a pin 2 of the isolating switch power supply U8 are respectively connected with a pin 1 and a pin 2 of a socket J1, a pin 1 of the voltage stabilizing chip 1117A is also connected with a pin J1.1 of a chip SIM900A, a capacitor C4 is connected between the pin 1 and the pin 4 of the voltage stabilizing chip 1117A, a capacitor C3 is connected in parallel with the capacitor C4, the capacitor C3 is connected with a light emitting diode and a resistor R1 which are mutually connected in series, the pin 4 of the voltage stabilizing chip 1117A is respectively connected with a pin 4, a pin 5, a pin 13 and a pin 9 of a chip AMG8833, the pin 4 of the voltage stabilizing chip 1117A is also connected with a capacitor C11, a capacitor C13 and a capacitor C14 which are mutually connected in parallel, and the pin 4 of the voltage stabilizing chip 1117A is connected with a capacitor C3, a pin 1, a pin 3 and a pin 30.

In the utility model, a chip AMG8833 is connected with a chip STM32 through an IIC interface, a chip ESP8266 is connected with the chip STM32 through a GPIO, a chip SIM900A is connected with the chip STM32 through a serial port, sensing data of the chip AMG8833 is transmitted to the STM32 through the IIC interface for data processing and identification, and an alarm signal is sent to the ESP8266 and SIM900A chips for alarm and notification.

The whole set of device is arranged at the middle position of the top of the room, and the height of the whole set of device is about 3-3.5 meters from the ground. The intelligent alarm system has the advantages that large shielding is avoided between the intelligent alarm system and the detected object, the AMG8833 array sensor in the equipment can detect three times per second, when someone falls down in a room, the AMG8833 array sensor can display a rectangular light-colored array, and the light-colored array is longer than the light-colored array in standing, as shown in fig. 5 and 6, so that the equipment recognizes falling behaviors and sends an alarm to the APP on the bound mobile phone, and meanwhile, the equipment also sends an alarm short message to the telephone number which is set, so that the indoor falling detection and alarm functions are realized, one set of independent rooms can be installed, and the independent rooms can operate independently, and can be detected in a time sensing mode no matter the cared person is in an independent space.

As shown in fig. 4, CN1 and P1 are respectively connected to ESP8266 and STM32, and the upper computer performs programming on both to implement the fall detection and data uploading functions of the device to the server.

The following is a detailed description of the operation of the present utility model.

1. Embodiments of the device to detect fall behavior

(1) AMG8833 performs three data acquisitions per second, each data format being an 8 x 8 temperature data array

(2) The AMG8833 transmits the data to the STM32 in real time over the IIC interface.

(3) STM32 performs computation, feature extraction and data classification judgment on the data.

(4) The data calculation, feature extraction and data classification judgment and flow are as follows:

i. STM32 obtains each frame of array data and then carries out pixel point maximum variance operation on each frame of array data

ii. If the maximum variance value is greater than the threshold value, the extraction of data features, namely the number of abrupt frames Fs, the range maximum variance Vfm and the maximum threshold pixel number Nm, is started.

The abrupt frame number is defined as the number of frames used for the human body action to change greatly in a short time, the range maximum variance is defined as the value of the maximum variance pixel point in the interval of the human body action frame, the maximum threshold pixel number Nm, and the number of pixels with larger variances in 8 x 8 pixel points of each frame is compared in all data frames from the instant frame with larger variance to the current frame.

iii, carrying out KNN classification on the data with the extracted characteristics, judging whether a falling action occurs, returning to the step i if no falling occurs, and sending alarm information to the ESP8266 and the SIM900A if falling occurs

And iii, if the array data received by the sensor does not change or the change value is too small within a period of time after sending the alarm, judging that the falling person is still in a falling state, and sending alarm information to the ESP8266 and the SIM900A again.

2. Implementation mode for identifying falling behavior and sending short message to server and reminded mobile phone by device

(1) After judging the falling action, STM32 sends alarm information to ESP8266 through serial port 1 and sends alarm information to SIM900A through serial port 2. The ESP8266 transmits data to the Ali cloud server through the home network and the router connection external network in the MQTT protocol.

(2) After receiving the alarm information, the SIM900A sends an alarm short message to the mobile phone number set in the program.

(3) And after receiving the fall alarm, the Alicloud server sends the alarm and the notification to a mobile visualization development program (WeChat applet, mobile APP or web visualization tool station) generated by a platform of the Alicloud server.

Based on the infrared sensors in the environment sensor, the sensor array data are collected by the singlechip and classified, so that the falling behavior is identified and judged. If the falling action is identified, the equipment is connected with the internet of things platform through the ESP8266 chip, so that the alarm information can be received on the mobile phone APP, and meanwhile, the alarm short message can be sent to the appointed mobile phone through the GSM module, so that double alarm reminding is realized.

The device is simple to use, the falling behaviors can be detected all the time only by supplying power to the device and placing the device at a proper position, and operations such as wearing or charging are not needed; the privacy protection is strong, the resolution ratio and the characteristics of the sensor can acquire the privacy information of the user as little as possible on the premise of normally realizing the functions, and the possibility of privacy disclosure is avoided from the source; the function implementation accuracy is high, multiple verification is adopted on analysis of falling behaviors, including human body detection, action classification, timing verification and long-time residence secondary alarm after falling, so that high-accuracy identification detection is realized, and misjudgment and false alarm are avoided.

The present utility model is not limited to the above-described embodiments, and based on the technical solutions of the present disclosure, those skilled in the art may make some substitutions and modifications to some technical features thereof without creative efforts, which are all within the scope of the present utility model.

Claims (4)

1. Human body fall detection device based on infrared array, its characterized in that: the system comprises an infrared array sensor, wherein the infrared array sensor is connected with a singlechip, the singlechip is respectively connected with an Internet of things module and a GSM module, and the infrared array sensor, the singlechip, the Internet of things module and the GSM module are all connected with a power supply module;

the singlechip includes chip STM32, infrared array sensor includes chip AMG8833, the No. 2 foot of chip STM32 links to each other with the 93 foot of chip STM32, the No. 3 foot of chip STM32 links to each other with the 92 foot of chip STM32, the 6 foot ground connection of chip STM32, the 13 foot and the 9 foot of chip STM32 link to each other with power module, the 10 foot of chip STM32 links to each other with the 12 foot of chip STM32 through resistance R4, electric capacity C9 and C10, electric capacity C9 one end links to each other with electric capacity C7 one end ground connection, the electric capacity C7 other end links to each other with the 9 foot of chip STM32, electric capacity C7 connects in parallel has electric capacity C8.

2. A personal fall detection device based on an infrared array as claimed in claim 1, wherein: the internet of things module includes chip ESP8266, and No. 1 foot, no. 3 foot, no. 4 foot, no. 29 foot and No. 30 foot of chip ESP8266 link to each other with power module, and No. 2 foot and the antenna of chip ESP8266 link to each other, and No. 7 foot of chip ESP8266 links to each other with 11 feet, no. 28 feet, no. 50 feet, no. 75 feet and No. 100 feet of chip STM32 through resistance R6, and No. 26 feet of chip ESP8266 link to each other with No. 25 feet of chip STM32, and No. 25 feet of chip ESP8266 link to each other with No. 26 feet of chip STM32.

3. A personal fall detection device based on an infrared array as claimed in claim 2, wherein: the GSM module comprises a chip SIM900A, wherein a J1.1 pin of the chip SIM900A is connected with the power supply module, a J1.2 pin of the chip SIM900A is connected with a J1.7 pin of the chip SIM900A and grounded, a J1.5 pin of the chip SIM900A is connected with a 68 pin of the chip STM32, and a J1.6 pin of the chip SIM900A is connected with a 69 pin of the chip STM32.

4. A personal fall detection device based on an infrared array as claimed in claim 3, wherein: the power supply module comprises a voltage stabilizing chip 1117A, a pin 1 and a pin 3 of the voltage stabilizing chip 1117A are connected with a pin 3 of an isolating switch power supply U8, a pin 4 of the isolating switch power supply U8 is grounded, a pin 1 and a pin 2 of the isolating switch power supply U8 are respectively connected with a pin 1 and a pin 2 of a socket J1, a pin 1 of the voltage stabilizing chip 1117A is also connected with a pin J1.1 of a chip SIM900A, a capacitor C4 is connected between the pin 1 and the pin 4 of the voltage stabilizing chip 1117A, a capacitor C3 is connected in parallel with the capacitor C4, the capacitor C3 is connected with a light emitting diode and a resistor R1 which are mutually connected in series, the pin 4 of the voltage stabilizing chip 1117A is respectively connected with a pin 4, a pin 5, a pin 13 and a pin 9 of a chip AMG8833, the pin 4 of the voltage stabilizing chip 1117A is also connected with a capacitor C11, a capacitor C13 and a capacitor C14 which are mutually connected in parallel, and the pin 4 of the voltage stabilizing chip 1117A is connected with a capacitor C3, a pin 1, a pin 3 and a pin 30.