Old people falling detection system

Abstract

The invention discloses an old people tumble detection system which is characterized in that human body acceleration data, human body inclination angle data and plantar pressure data are acquired and input into a judgment module, physical sign data are acquired and input into the judgment module, a tumble judgment result is acquired and input into an inflation module, and the inflation module is used for inflating the inflation module. According to the acceleration data, the human body inclination angle data, the plantar pressure data and the physical sign data, performing early warning judgment, obtaining early warning judgment, inputting an early warning judgment result into an inflation module, receiving a falling judgment result and the early warning judgment result, and performing inflation according to the received falling judgment result and the early warning judgment result. Accurate detection, personalized early warning, dynamic adjustment and intelligent inflation are integrated, through multi-sensor integration, motion data are analyzed in real time, tumble is accurately judged, inflation protection is carried out instantly, meanwhile, the tumble risk is evaluated according to physical sign data, multi-layer early warning is provided, the tumble risk of old people is effectively reduced, and safety guarantee is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of fall detection, and in particular to a fall detection system for the elderly. Background Art

[0002] Falls among the elderly are a global health problem, and their consequences are often very serious and may even lead to death. Therefore, it is particularly important to develop an efficient, accurate and reliable fall detection system for the elderly.

[0003] At present, the Chinese invention patent with application number CN202110397698.5 discloses an intelligent human fall detection system and method. The intelligent human fall detection system of this invention includes a regional video acquisition sensor, a regional human detection model, a regional human fall detection model and an alarm unit. The regional video acquisition sensor is used to collect video images of scenes in the region and send them to the regional human detection model; the regional human detection model is used to receive video images sent by the regional video acquisition sensor; the regional human fall detection model is used to detect whether a human body falls according to the video image, and if a human body falls, the human body fall information is sent to the alarm unit; the alarm unit is used to receive the human body fall information sent by the regional human body fall detection model and issue an alarm. Summary of the invention

[0004] The technical problem solved by the present invention is that it is difficult in the above-mentioned technology to provide graded warnings and reminders to the human body through acceleration data, human body inclination data, plantar pressure data and vital sign data, and it is difficult to improve the accuracy and reliability of fall detection.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] An elderly fall detection system includes a sensing module, a collection module, a judgment module and an inflation module;

[0007] The sensing module is used to obtain human body acceleration data, human body inclination data and plantar pressure data, and input the human body acceleration data, human body inclination data and plantar pressure data into the judgment module;

[0008] The acquisition module is used to collect vital sign data and input the vital sign data into the judgment module;

[0009] The judgment module is used to make a fall judgment based on the human body acceleration data, the human body inclination data and the plantar pressure data, obtain a fall judgment result, and input the fall judgment result into the inflation module, make an early warning judgment based on the acceleration data, the human body inclination data, the plantar pressure data and the vital sign data, obtain an early warning judgment, and input the early warning judgment result into the inflation module;

[0010] The inflation module is used for receiving the fall judgment result and the early warning judgment result, and inflating according to the received fall judgment result and the early warning judgment result.

[0011] Preferably, the sensing module includes an acceleration sensor, a magnetic sensor and a pressure sensor;

[0012] The acceleration sensor is used to collect human body acceleration data;

[0013] The magnetic sensor is used to collect human body inclination data;

[0014] The pressure sensor is used to collect plantar pressure data.

[0015] Preferably, the judgment module includes a fall judgment unit and a warning judgment unit;

[0016] The fall judgment unit is used to receive human body acceleration data, human body inclination data and plantar pressure data, and use the human body acceleration data as a first judgment, the human body inclination data as a second judgment, and the plantar pressure as a third judgment;

[0017] If the human body acceleration data is greater than the predetermined acceleration threshold, it means that the first judgment is met and a second judgment is performed;

[0018] If the human body inclination angle data is greater than the predetermined inclination angle threshold, it means that the double judgment is met and a triple judgment is performed;

[0019] If the plantar pressure data is greater than a predetermined pressure threshold, it means that the triple judgment is met, and a fall judgment result is output, where the fall judgment result is a fall instruction or a no-fall instruction.

[0020] Preferably, the physical sign data include weight data, age, medical history and living habits.

[0021] Preferably, the acquisition module is also used to obtain the pressure threshold of the human body based on the weight data. The pressure sensor collects the weight data in real time, and the weight data is input into a weight and pressure threshold correspondence table pre-built in the acquisition module to obtain a new pressure threshold of the human body in real time. The new pressure threshold of the human body is input into the fall judgment unit to update the pressure threshold in the judgment unit.

[0022] Preferably, the acquisition module further includes a monitoring unit, which is used to obtain information about the surrounding environment, determine whether there is a collision risk between the human body and the surrounding environment, and if there is a collision risk, generate a warning signal, and input the warning signal into the inflation module;

[0023] The monitoring unit takes real-time photos of the surrounding environment, collects image information of the human body, and inputs the image information into a three-dimensional space model to obtain three-dimensional coordinate information, and calibrates reference points, wherein the reference points include a fixed reference point group and a mobile reference point group, wherein the fixed reference point group is set at the human body's feet, calves, thighs, waist, chest and shoulders, and the mobile reference point group is set on the surrounding environment and is updated as the surrounding environment changes;

[0024] Acquire and construct human body position information according to the three-dimensional coordinate information of the fixed reference point group, adjust the human body position information according to the time axis, acquire human body movement trajectory information, judge life habits according to the human body movement trajectory information, and collect data exceeding the initial acceleration threshold, inclination threshold and pressure threshold in the changes of acceleration data, human body inclination data and plantar pressure data in life habits, and acquire the corresponding time period information of the data exceeding the initial acceleration threshold, inclination threshold and pressure threshold in the changes of acceleration data, human body inclination data and plantar pressure data;

[0025] If the acceleration data, human body inclination data and plantar pressure data exceed the updated acceleration threshold, inclination threshold and pressure threshold in the corresponding time period information, a no-fall instruction is output, indicating that there is no fall.

[0026] Preferably, the early warning judgment unit is used to analyze the weight data:

[0027] If the weight data is greater than the predetermined weight threshold, the output is a high-risk weight type for falling;

[0028] Analysis of age:

[0029] If the age is greater than the predetermined age threshold, the output is a high-risk age type for falling;

[0030] According to the predetermined disease type library, find the types of diseases that affect balance and coordination in the past medical history:

[0031] If the past medical history includes diseases that affect balance and coordination, the output is a high-risk disease type for falling;

[0032] Perform text segmentation on living habits to obtain a vocabulary phrase set. Based on the predetermined target keyword list, use a string matching algorithm to search whether the vocabulary phrase set contains any item in the target keyword list:

[0033] If the vocabulary phrase set contains any item in the target keyword list, the output is a habit type with a high risk of falling;

[0034] The high-fall risk weight type, high-fall risk age type, high-fall risk disease type and high-fall risk habit type are combined and output as a high-fall risk physical sign type.

[0035] Preferably, the early warning judgment unit is further used to analyze the acceleration data and calculate the acceleration change rate between adjacent accelerations:

[0036] If the acceleration change rate is greater than the predetermined acceleration change rate threshold, the output is the acceleration mutation type;

[0037] Analyze the human body inclination data and calculate the inclination change rate between adjacent human body inclinations:

[0038] If the tilt angle change rate is greater than the predetermined tilt angle change rate threshold, the output is a tilt angle mutation type;

[0039] Analyze the plantar pressure data and calculate the pressure change rate between adjacent plantar pressures:

[0040] If the pressure change rate is greater than the predetermined pressure change rate threshold, the output is a pressure mutation type;

[0041] The acceleration mutation type, the inclination mutation type and the pressure mutation type are combined and output as the human body data mutation type;

[0042] If the physical sign data meets any of the high-risk physical sign types for falling, and the analysis results of the acceleration data, human body inclination data, and plantar pressure data meet any of the human body data mutation types, the output is a first-level fall warning signal;

[0043] If the physical sign data meets any one of the high-risk physical sign types for falling, and the analysis results of the acceleration data, human body inclination data, and plantar pressure data meet any two of the human body data mutation types, the output is a secondary fall warning signal;

[0044] If the analysis results of the acceleration data, the human body inclination data and the plantar pressure data meet all the items in the human body data mutation type, the output is a level 3 fall warning signal;

[0045] The first-level fall warning signal, the second-level fall warning signal and the third-level fall warning signal are combined and output as the warning judgment result.

[0046] Preferably, the inflation module includes a result analysis unit and an inflation start unit;

[0047] The result analysis unit is used for analyzing the fall judgment result and the early warning judgment result, and outputting the inflation intensity signal;

[0048] The inflation start unit is used to receive the inflation intensity signal, open the electric valve, and automatically inflate the inflatable airbag according to the inflation intensity signal.

[0049] Preferably, the analysis of the fall judgment result and the warning judgment result is as follows:

[0050] If the fall judgment result is a no-fall instruction and the warning judgment result is a first-level fall warning signal, a first inflation intensity signal is output;

[0051] If the fall judgment result is a no-fall instruction and the warning judgment result is a second-level fall warning signal, a second inflation intensity signal is output;

[0052] If the fall judgment result is a no-fall instruction and the warning judgment result is a level 3 fall warning signal, a third inflation intensity signal is output;

[0053] If the fall judgment result is a fall instruction and the warning judgment result is a level 3 fall warning signal, a fourth inflation intensity signal is output;

[0054] The first inflation intensity signal, the second inflation intensity signal, the third inflation intensity signal and the fourth inflation intensity signal are combined and output as an inflation intensity signal.

[0055] The beneficial effects of the present invention are as follows: the present invention integrates precise detection, personalized early warning, dynamic adjustment and intelligent inflation. Through multi-sensor integration, it analyzes motion data in real time, accurately determines falls and immediately inflates for protection. At the same time, it evaluates the risk of falls based on physical sign data, provides multi-layer early warnings, effectively reduces the risk of falls for the elderly, and improves safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 A schematic diagram of the basic flow of a fall detection system for the elderly provided by one embodiment of the present invention; DETAILED DESCRIPTION

[0057] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0058] Example 1

[0059] Reference Figure 1 , which is an embodiment of the present invention, provides a fall detection system for the elderly, including a sensing module, a collection module, a judgment module and an inflation module.

[0060] The sensing module is used to obtain human body acceleration data, human body inclination data and plantar pressure data, and input the human body acceleration data, human body inclination data and plantar pressure data into the judgment module.

[0061] The collection module is used to collect vital sign data and input the vital sign data into the judgment module.

[0062] The judgment module is used to make a fall judgment based on the human body acceleration data, the human body inclination data and the plantar pressure data, obtain the fall judgment result, and input the fall judgment result into the inflation module, make an early warning judgment based on the acceleration data, the human body inclination data, the plantar pressure data and the vital sign data, obtain the early warning judgment, and input the early warning judgment result into the inflation module.

[0063] The inflation module is used for receiving the fall judgment result and the early warning judgment result, and inflating according to the received fall judgment result and the early warning judgment result.

[0064] The sensing module includes an acceleration sensor, a magnetic sensor and a pressure sensor.

[0065] The acceleration sensor is used to collect human body acceleration data.

[0066] The magnetic sensor is used to collect human body inclination data.

[0067] The pressure sensor is used to collect plantar pressure data.

[0068] The sensing module integrates acceleration sensors, magnetic sensors and pressure sensors to achieve comprehensive monitoring of the movement status of the elderly, and accurately obtains the human body's acceleration data, inclination data and plantar pressure data. These data are important bases for judging whether the elderly have fallen. The efficient and accurate operation of the sensing module provides reliable data support for subsequent fall judgments and early warning judgments, ensuring the accuracy and reliability of the entire fall detection system.

[0069] Physical data include weight data, age, medical history and lifestyle habits.

[0070] The judgment module includes a fall judgment unit and a warning judgment unit.

[0071] The fall judgment unit is used to receive human body acceleration data, human body inclination data and plantar pressure data, and uses the human body acceleration data as the first judgment, the human body inclination data as the second judgment, and the plantar pressure as the third judgment.

[0072] If the human body acceleration data is greater than a predetermined acceleration threshold, it means that the first judgment is met and a second judgment is performed.

[0073] If the human body inclination angle data is greater than a predetermined inclination angle threshold, it means that the double judgment is met and a triple judgment is performed.

[0074] If the plantar pressure data is greater than a predetermined pressure threshold, it means that the triple judgment is met, and a fall judgment result is output, where the fall judgment result is a fall instruction or a no-fall instruction.

[0075] The fall judgment unit uses a triple judgment mechanism to accurately analyze the movement status of the elderly. When an abnormal movement pattern is detected, it can quickly make a fall judgment and output the corresponding fall command or no fall command. This instant and accurate judgment provides strong support for subsequent emergency response.

[0076] The early warning judgment unit is used to analyze the weight data:

[0077] If the weight data is greater than a predetermined weight threshold, the output is a high-risk weight type for falling.

[0078] Analysis of age:

[0079] If the age is greater than a predetermined age threshold, the output is a high-risk age type for falling.

[0080] According to the predetermined disease type library, find the types of diseases that affect balance and coordination in the past medical history:

[0081] If the past medical history includes disease types that affect balance and coordination, the output is a disease type with a high risk of falling.

[0082] Perform text segmentation on living habits to obtain a vocabulary phrase set. Based on the predetermined target keyword list, use a string matching algorithm to search whether the vocabulary phrase set contains any item in the target keyword list:

[0083] If the vocabulary phrase set contains any item in the target keyword list, the output is a habit type with a high risk of falling.

[0084] The high-fall risk weight type, high-fall risk age type, high-fall risk disease type and high-fall risk habit type are combined and output as a high-fall risk physical sign type.

[0085] The early warning judgment unit is also used to analyze the acceleration data and calculate the acceleration change rate between adjacent accelerations:

[0086] If the acceleration change rate is greater than the preset acceleration change rate threshold, the output is the acceleration mutation type.

[0087] Analyze the human body inclination data and calculate the inclination change rate between adjacent human body inclinations:

[0088] If the tilt angle change rate is greater than a predetermined tilt angle change rate threshold, the output is a tilt angle mutation type.

[0089] Analyze the plantar pressure data and calculate the pressure change rate between adjacent plantar pressures:

[0090] If the pressure change rate is greater than a predetermined pressure change rate threshold, the output is a pressure mutation type.

[0091] The acceleration mutation type, inclination mutation type and pressure mutation type are combined and output as the human body data mutation type.

[0092] If the physical sign data meets any of the high-risk physical sign types for falling, and the analysis results of the acceleration data, human body inclination data and plantar pressure data meet any of the human body data mutation types, the output is a first-level fall warning signal.

[0093] If the physical sign data meets any of the high-risk fall sign types, and the analysis results of the acceleration data, human body inclination data and plantar pressure data meet any two of the human body data mutation types, the output is a secondary fall warning signal.

[0094] If the analysis results of the acceleration data, the human body inclination data and the plantar pressure data meet all the items in the human body data mutation type, the output is a level 3 fall warning signal.

[0095] The first-level fall warning signal, the second-level fall warning signal and the third-level fall warning signal are combined and output as the warning judgment result.

[0096] The early warning judgment unit focuses on the comprehensive assessment of the risk of falls for the elderly. By analyzing vital signs data such as weight, age, medical history and living habits, it identifies individuals with high risk of falls. At the same time, the unit also monitors the rate of change of motion data such as acceleration, inclination and plantar pressure in real time. Once abnormal fluctuations are detected, different levels of fall warning signals are issued in combination with the vital signs data. This multi-level early warning mechanism helps to take preventive measures in advance and reduce the occurrence of falls.

[0097] The judgment module is the core of the elderly fall detection system. It integrates the fall judgment unit and the early warning judgment unit, realizing the accurate identification and risk assessment of elderly fall events. By comprehensively analyzing the real-time data and vital signs of the human body, the judgment module can quickly make a fall judgment and send out a warning signal in advance when necessary, effectively reducing the risk and consequences of falls for the elderly.

[0098] The acquisition module is also used to obtain the pressure threshold of the human body based on the weight data. The pressure sensor collects the weight data in real time, inputs the weight data into the pre-built weight and pressure threshold correspondence table in the acquisition module, obtains the new pressure threshold of the human body in real time, inputs the new pressure threshold of the human body into the fall judgment unit, and updates the pressure threshold in the judgment unit.

[0099] The acquisition module also includes a monitoring unit, which is used to obtain information about the surrounding environment and determine whether there is a risk of collision between the human body and the surrounding environment. If there is a risk of collision, a warning signal is generated and the warning signal is input into the inflation module.

[0100] The monitoring unit takes real-time photos of the surrounding environment, collects image information of the human body's surroundings, inputs the image information into a three-dimensional space model, obtains three-dimensional coordinate information, and calibrates reference points. The reference points include a fixed reference point group and a mobile reference point group. The fixed reference point group is set on the human body's feet, calves, thighs, waist, chest and shoulders, and the mobile reference point group is set on the surrounding environment and updated as the surrounding environment changes.

[0101] The human body position information is acquired and constructed based on the three-dimensional coordinate information of the fixed reference point group, and the human body position information is adjusted according to the time axis to acquire human body movement trajectory information, judge life habits based on the human body movement trajectory information, and collect data that exceeds the initial acceleration threshold, inclination threshold and pressure threshold in the changes of acceleration data, human body inclination data and plantar pressure data in life habits, and acquire the time period information corresponding to the data that exceeds the initial acceleration threshold, inclination threshold and pressure threshold in the changes of acceleration data, human body inclination data and plantar pressure data.

[0102] If the acceleration data, human body inclination data and plantar pressure data exceed the updated acceleration threshold, inclination threshold and pressure threshold in the corresponding time period information, a no-fall instruction is output, indicating that there is no fall.

[0103] The acquisition module collects weight data in real time and uses the corresponding relationship table between weight and pressure threshold to dynamically update the pressure threshold in the fall judgment unit. This personalized and dynamic adjustment method can more accurately adapt to the needs of elderly people with different weights and improve the accuracy and reliability of fall judgment. The monitoring unit realizes real-time monitoring of the collision risk between the human body and the surrounding environment by taking real-time photos of the surrounding environment and building a three-dimensional space model. Once a potential collision risk is found, an early warning signal is immediately generated and transmitted to the inflation module so that inflation protection measures can be taken in time, effectively avoiding fall accidents caused by environmental collision. The monitoring unit not only pays attention to the immediate collision risk, but also deeply explores the living habits of the elderly through the analysis of human movement trajectory information. By identifying abnormal changes in acceleration, inclination and plantar pressure data in living habits, and combining the time axis for detailed analysis, the system can more comprehensively evaluate the fall risk of the elderly. This fall risk assessment based on living habits provides strong support for the formulation of personalized prevention strategies. On the basis of comprehensive consideration of dynamic pressure thresholds, environmental collision risk warnings and living habits analysis, the system further optimizes fall judgment. By comparing whether the acceleration, inclination and plantar pressure data exceed the updated threshold within a specific time period, the system can more accurately determine whether the elderly person has actually fallen, avoiding misjudgment and missed judgment.

[0104] The inflation module includes a result analysis unit and an inflation starting unit.

[0105] The result analysis unit is used for analyzing the fall judgment result and the early warning judgment result, and outputting the inflation intensity signal.

[0106] The fall judgment results and warning judgment results are analyzed as follows:

[0107] If the fall judgment result is a no-fall instruction and the warning judgment result is a first-level fall warning signal, a first inflation intensity signal is output;

[0108] If the fall judgment result is a no-fall instruction and the warning judgment result is a second-level fall warning signal, a second inflation intensity signal is output;

[0109] If the fall judgment result is a no-fall instruction and the warning judgment result is a level 3 fall warning signal, a third inflation intensity signal is output;

[0110] If the fall judgment result is a fall instruction and the warning judgment result is a level 3 fall warning signal, a fourth inflation intensity signal is output;

[0111] The first inflation intensity signal, the second inflation intensity signal, the third inflation intensity signal and the fourth inflation intensity signal are combined and output as an inflation intensity signal.

[0112] The result analysis unit is responsible for receiving the fall judgment results and the warning judgment results, and using these data comprehensively for analysis and judgment. According to different combinations of judgment results, it can output different levels of inflation intensity signals. This graded output method not only takes into account the severity of the fall risk, but also takes into account the actual needs of the elderly in different situations, ensuring the targetedness and effectiveness of inflation protection.

[0113] The inflation start unit is used to receive the inflation intensity signal, open the electric valve, and automatically inflate the inflatable airbag according to the inflation intensity signal.

[0114] The inflation start unit works closely with the result analysis unit. Once the inflation strength signal from the result analysis unit is received, the unit will respond immediately, open the electric valve, and automatically inflate the airbag according to the signal instructions. This fast and accurate response mechanism ensures that the inflation protection can be started in the shortest time and provide timely support and protection for the elderly. At the same time, it can also adjust the inflation volume according to the different inflation strength signals to meet different levels of protection needs.

[0115] As the key executive part of the elderly fall detection system, the inflation module is designed to take protective measures quickly and accurately based on the fall judgment results and early warning judgment results. The result analysis unit conducts in-depth analysis of the judgment results and outputs the corresponding inflation intensity signal accordingly. The inflation start unit controls the opening of the electric valve according to these signals and automatically inflates the inflatable airbag, thereby providing the necessary support and protection for the elderly. This ability of instant response and precise control greatly reduces the possible injuries that the elderly may suffer after falling, and improves practicality and safety.

[0116] By integrating accelerometers, magnetic sensors and pressure sensors, it is possible to comprehensively and accurately obtain the movement status data of the elderly, including acceleration, inclination angle and plantar pressure, so as to accurately detect falls. Once a fall is detected, it can respond immediately and quickly inflate the inflatable module to provide the necessary support and protection for the elderly and reduce the risk of injury. It not only pays attention to the immediate fall events, but also collects and analyzes the elderly's weight, age, medical history and living habits and other physical data to provide personalized fall risk warnings. This early warning mechanism can identify potential risk factors in advance, help the elderly and their families or caregivers take preventive measures to avoid the occurrence of falls, and can adjust the plantar pressure threshold in real time according to the weight changes of the elderly to ensure the accuracy of fall detection. At the same time, through real-time monitoring of the surrounding environment and three-dimensional space modeling, the human body position information and movement trajectory can be dynamically adjusted to further improve the accuracy and reliability of fall detection. The present invention establishes a multi-level fall warning system, including primary, secondary and tertiary warning signals. This hierarchical warning mechanism can take corresponding measures according to different risk levels, which not only avoids excessive intervention, but also ensures timely and effective response. The inflation module automatically adjusts the inflation intensity according to the fall judgment result and the warning judgment result, providing personalized protection for the elderly. Before or at the moment of falling, the inflation module can be quickly inflated to provide support for the elderly and reduce the impact force; after falling, the inflation module can continue to maintain the inflation state to ensure that the elderly are in a safe and comfortable posture, waiting for rescue, and privacy protection and user experience are fully considered in the design. The non-contact monitoring method avoids infringement of the privacy of the elderly; at the same time, it is easy to operate and responds quickly, which can minimize the impact on the daily life of the elderly and improve the user experience.

[0117] It should be appreciated that embodiments of the present invention may be implemented or enforced by a combination of computer hardware and software, or by computer instructions stored in a non-transitory computer-readable memory. The method may be implemented in a computer program using standard programming techniques - including a non-transitory computer-readable storage medium configured with a computer program, wherein the storage medium so configured causes the computer to operate in a specific and predefined manner - according to the methods and drawings described in the specific embodiments. Each program may be implemented in a high-level procedural or object-oriented programming language to communicate with a computer system. However, if desired, the program may be implemented in an assembly or machine language. In any case, the language may be a compiled or interpreted language. In addition, the program may be run on a programmed ASIC for this purpose.

[0118] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A fall detection system for the elderly, characterized in that: It includes a sensing module, a collection module, a judgment module and an inflation module; The sensing module is used to obtain human body acceleration data, human body inclination data and plantar pressure data, and input the human body acceleration data, human body inclination data and plantar pressure data into the judgment module; The acquisition module is used to collect vital sign data and input the vital sign data into the judgment module; The judgment module is used to make a fall judgment based on the human body acceleration data, the human body inclination data and the plantar pressure data, obtain a fall judgment result, and input the fall judgment result into the inflation module, make an early warning judgment based on the acceleration data, the human body inclination data, the plantar pressure data and the vital sign data, obtain an early warning judgment, and input the early warning judgment result into the inflation module; The inflation module is used for receiving the fall judgment result and the early warning judgment result, and inflating according to the received fall judgment result and the early warning judgment result.

2. The elderly fall detection system as claimed in claim 1, characterized in that: The sensing module includes an acceleration sensor, a magnetic sensor and a pressure sensor; The acceleration sensor is used to collect human body acceleration data; The magnetic sensor is used to collect human body inclination data; The pressure sensor is used to collect plantar pressure data.

3. The elderly fall detection system as claimed in claim 2, characterized in that: The judgment module includes a fall judgment unit and a warning judgment unit; The fall judgment unit is used to receive human body acceleration data, human body inclination data and plantar pressure data, and use the human body acceleration data as a first judgment, the human body inclination data as a second judgment, and the plantar pressure as a third judgment; If the human body acceleration data is greater than the predetermined acceleration threshold, it means that the first judgment is met and a second judgment is performed; If the human body inclination angle data is greater than the predetermined inclination angle threshold, it means that the double judgment is met and a triple judgment is performed; If the plantar pressure data is greater than a predetermined pressure threshold, it means that the triple judgment is met, and a fall judgment result is output, where the fall judgment result is a fall instruction or a no-fall instruction.

4. The elderly fall detection system as claimed in claim 3, characterized in that: The physical sign data include weight data, age, past medical history and living habits.

5. The elderly fall detection system as claimed in claim 4, characterized in that: The acquisition module is also used to obtain the pressure threshold of the human body based on the weight data. The pressure sensor collects the weight data in real time, inputs the weight data into the acquisition module to pre-build a weight and pressure threshold correspondence table, obtains the new pressure threshold of the human body in real time, inputs the new pressure threshold of the human body into the fall judgment unit, and updates the pressure threshold in the judgment unit.

6. The elderly fall detection system as claimed in claim 5, characterized in that: The acquisition module also includes a monitoring unit, which is used to obtain information about the surrounding environment, determine whether there is a risk of collision between the human body and the surrounding environment, and if there is a risk of collision, generate a warning signal, and input the warning signal into the inflation module; The monitoring unit takes real-time photos of the surrounding environment, collects image information of the human body, and inputs the image information into a three-dimensional space model to obtain three-dimensional coordinate information, and calibrates reference points, wherein the reference points include a fixed reference point group and a mobile reference point group, wherein the fixed reference point group is set at the human body's feet, calves, thighs, waist, chest and shoulders, and the mobile reference point group is set on the surrounding environment and is updated as the surrounding environment changes; Acquire and construct human body position information according to the three-dimensional coordinate information of the fixed reference point group, adjust the human body position information according to the time axis, acquire human body movement trajectory information, judge life habits according to the human body movement trajectory information, and collect data exceeding the initial acceleration threshold, inclination threshold and pressure threshold in the changes of acceleration data, human body inclination data and plantar pressure data in life habits, and acquire the corresponding time period information of the data exceeding the initial acceleration threshold, inclination threshold and pressure threshold in the changes of acceleration data, human body inclination data and plantar pressure data; If the acceleration data, human body inclination data and plantar pressure data exceed the updated acceleration threshold, inclination threshold and pressure threshold in the corresponding time period information, a no-fall instruction is output, indicating that there is no fall.

7. The elderly fall detection system as claimed in claim 6, characterized in that: The early warning judgment unit is used to analyze the weight data: If the weight data is greater than the predetermined weight threshold, the output is a high-risk weight type for falling; Analysis of age: If the age is greater than the predetermined age threshold, the output is a high-risk age type for falling; According to the predetermined disease type library, find the types of diseases that affect balance and coordination in the past medical history: If the past medical history includes diseases that affect balance and coordination, the output is a high-risk disease type for falling; Perform text segmentation on living habits to obtain a vocabulary phrase set. Based on the predetermined target keyword list, use a string matching algorithm to search whether the vocabulary phrase set contains any item in the target keyword list: If the vocabulary phrase set contains any item in the target keyword list, the output is a habit type with a high risk of falling; The high-fall risk weight type, high-fall risk age type, high-fall risk disease type and high-fall risk habit type are combined and output as a high-fall risk physical sign type.

8. The elderly fall detection system as claimed in claim 7, characterized in that: The early warning judgment unit is also used to analyze the acceleration data and calculate the acceleration change rate between adjacent accelerations: If the acceleration change rate is greater than the predetermined acceleration change rate threshold, the output is the acceleration mutation type; Analyze the human body inclination data and calculate the inclination change rate between adjacent human body inclinations: If the tilt angle change rate is greater than the predetermined tilt angle change rate threshold, the output is a tilt angle mutation type; Analyze the plantar pressure data and calculate the pressure change rate between adjacent plantar pressures: If the pressure change rate is greater than the predetermined pressure change rate threshold, the output is a pressure mutation type; The acceleration mutation type, the inclination mutation type and the pressure mutation type are combined and output as the human body data mutation type; If the physical sign data meets any of the high-risk physical sign types for falling, and the analysis results of the acceleration data, human body inclination data, and plantar pressure data meet any of the human body data mutation types, the output is a first-level fall warning signal; If the physical sign data meets any one of the high-risk physical sign types for falling, and the analysis results of the acceleration data, human body inclination data, and plantar pressure data meet any two of the human body data mutation types, the output is a secondary fall warning signal; If the analysis results of the acceleration data, the human body inclination data and the plantar pressure data meet all the items in the human body data mutation type, the output is a level 3 fall warning signal; The first-level fall warning signal, the second-level fall warning signal and the third-level fall warning signal are combined and output as the warning judgment result.

9. The elderly fall detection system as claimed in claim 8, characterized in that: The inflation module includes a result analysis unit and an inflation start unit; The result analysis unit is used for analyzing the fall judgment result and the early warning judgment result, and outputting the inflation intensity signal; The inflation start unit is used to receive the inflation intensity signal, open the electric valve, and automatically inflate the inflatable airbag according to the inflation intensity signal.

10. The elderly fall detection system as claimed in claim 9, characterized in that: The analysis of the fall judgment result and the early warning judgment result is as follows: If the fall judgment result is a no-fall instruction and the warning judgment result is a first-level fall warning signal, a first inflation intensity signal is output; If the fall judgment result is a no-fall instruction and the warning judgment result is a second-level fall warning signal, a second inflation intensity signal is output; If the fall judgment result is a no-fall instruction and the warning judgment result is a level 3 fall warning signal, a third inflation intensity signal is output; If the fall judgment result is a fall instruction and the warning judgment result is a level 3 fall warning signal, a fourth inflation intensity signal is output; The first inflation intensity signal, the second inflation intensity signal, the third inflation intensity signal and the fourth inflation intensity signal are combined and output as an inflation intensity signal.

Images