Millimeter wave radar-based fall detection offline voice confirmation method and device, and storage medium device

Abstract

The invention discloses a millimeter wave radar-based fall detection offline voice confirmation method and device, and a storage medium device, and relates to the technical field of fall detection, and the method specifically comprises the following steps: (a) millimeter wave radar monitoring; after being started, the millimeter-wave radar module continuously emits millimeter-wave signals and receives reflected signals, the signals of the millimeter-wave radar module are processed through a microprocessor signal processing algorithm, height information of a target is obtained, the system conducts real-time monitoring through the height information obtained through calculation, and the current height Gcurrent and the dynamic height threshold Hth and Hcurrent lt are compared; and when Hth is greater than Hth, the system preliminarily judges that the target may fall down. The system has the characteristics of good privacy protection, high detection sensitivity, high accuracy and strong environmental adaptability, can realize the functions of long-term continuous real-time human body existence, tumble, static residence and voice recognition help seeking, and reduces the situation that the elderly living alone cannot be rescued in time after tumble.

Description

Technical Field

[0001] The present invention relates to the technical field of fall detection, and in particular to an offline voice confirmation method, device and storage medium for fall detection based on millimeter wave radar. Background Art

[0002] Falls are the leading cause of fatal injuries in the elderly, such as fractures. Falls can lead to loss of consciousness due to impact, or conversely, loss of consciousness can lead to falls, which in severe cases can result in death. Therefore, the risk of falls is a critical issue in today's aging society. According to the World Health Organization, 28% to 35% of the population aged 65 and over experience at least one fall per year. These falls account for at least 50% of hospitalizations in the elderly, as well as approximately 40% of deaths from non-natural causes. In addition, medical analysis of the damage caused by falls shows that it is highly dependent on reaction and rescue time. That is, the earlier the fall is detected, the lower the chance of death due to secondary injuries.

[0003] In the existing technology, it is impossible to continuously monitor the human body's falls, stationary stays and other states in real time and recognize voice requests for help for a long time, resulting in frequent situations where elderly people living alone cannot get timely help after falling. To this end, we propose an offline voice confirmation method, device and storage medium based on millimeter-wave radar fall detection to solve the above problems. Summary of the invention

[0004] The object of the present invention is to provide a method, device and storage medium for offline voice confirmation based on millimeter wave radar fall detection to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: an offline voice confirmation method based on millimeter wave radar fall detection, which specifically includes the following steps:

[0006] Step (a): Millimeter-wave radar monitoring: After being turned on, the millimeter-wave radar module continuously transmits millimeter-wave signals and receives reflected signals. The millimeter-wave radar module signal is processed by the microprocessor signal processing algorithm to obtain the target's height information. The system performs real-time monitoring based on the calculated height information and compares the current height H current and dynamic height threshold H th , when H current <H th When the system initially determines that the target may fall;

[0007] Step (b): voice recognition confirmation; after the height information is lower than the set threshold, the system activates the microphone, starts to capture the audio signal of the surrounding environment, records the audio data, processes the audio signal through the offline voice interaction module recognition chip, and recognizes specific voice commands. The system has a built-in voice library that can recognize common help phrases, such as "I fell down", "Please help", etc. If the help signal is successfully recognized, the system further confirms that the target is indeed in a falling state;

[0008] Step (c): Multiple confirmation mechanism: To improve accuracy, the system is designed with a multiple confirmation mechanism. When the height information is lower than the threshold for multiple consecutive times and the voice recognition results are consistent, the system confirms that the target is in a falling state.

[0009] Step (d): Upload information; once the target falls, the system uploads relevant information, such as timestamp, user height, fall confirmation status, recognized voice commands, etc., to the cloud server through the communication module. The cloud server can receive and store data in real time for subsequent analysis and monitoring. The user or his relatives can access the cloud server at any time through a mobile device or computer to view the fall event record and related statistical data, thereby realizing remote monitoring.

[0010] An offline voice confirmation device based on millimeter wave radar fall detection, comprising:

[0011] Millimeter wave radar module; the millimeter wave radar module is used to send, receive and process millimeter wave signals, and store preprocessing programs for processing millimeter wave signals;

[0012] Microprocessor; the microprocessor is used to control the operation, calculation and storage of the module, process the millimeter wave signal, measure the parameters of the target object and determine the fall, and send the fall signal;

[0013] Offline voice interaction module; the offline voice interaction module is used to receive the fall signal sent by the microprocessor, interact with the user, and convert the user feedback information into warning information and help information;

[0014] A communication module, the communication module is used to receive the alarm information and help information commands of the offline voice interaction module, and send the alarm information and help information;

[0015] The microprocessor is respectively connected to the millimeter wave radar module, the offline voice interaction module, and the communication module. The communication module is connected to the cloud server. The cloud server is used to receive the alarm information and help information of the offline voice interaction module sent by the communication module.

[0016] Furthermore, the millimeter wave radar module includes:

[0017] Transmitting module; the transmitting module is used to transmit millimeter wave signals;

[0018] Receiving module; the receiving module is used to receive the millimeter wave echo signal emitted by the transmitting module and scattered and reflected by the target object;

[0019] Control unit module; the control unit module is used to control the receiving module to receive the millimeter wave echo signal and pre-process the millimeter wave echo signal;

[0020] A storage unit module, the storage unit module is used to store a preprocessing program for processing millimeter wave echo signals;

[0021] Wherein, the control unit module is respectively connected to the transmitting module, the receiving module and the storage unit module.

[0022] Furthermore, the microprocessor is used to control the operation, calculation and storage of the millimeter wave radar module, the offline voice interaction module and the communication module.

[0023] Furthermore, the millimeter wave signal processing specifically includes the following process: by processing the received millimeter wave multi-channel echo signal and the transmitted mixed intermediate frequency signal, using fast Fourier transform to process the intermediate frequency signal, and generate a preliminary spectrum diagram;

[0024] The target object parameter measurement specifically includes the following process: by analyzing the changes in various parameters such as the intensity frequency and phase of the received millimeter wave signal, some important parameters of the target object, such as distance, speed, azimuth, etc., can be obtained. The parameters can reflect the three-dimensional position information and height data of the target object in the detection area;

[0025] The target object fall judgment specifically includes the following process: calculating the acquired current height information and the fitted dynamic height threshold, and judging whether the moving target falls by comparing the two values.

[0026] Furthermore, the conversion into alarm information specifically includes the following process: the microprocessor sends the fall signal to the offline voice interaction module, the device broadcasts the fall confirmation message and the user feedback message through the speaker and microphone device, and within the specified time, according to the fall voice signal fed back by the user's shouting response, it is recognized by the voice recognition system and converted into a fall or release alarm message command, and then the corresponding alarm message is sent to the cloud server through the communication module;

[0027] The conversion into help information specifically includes the following process: when the user feels unwell, he can actively call for help, the device microphone collects the call voice signal, the voice recognition chip in the offline voice interaction module recognizes the rescue command, and the microprocessor then sends the rescue command to the cloud server through the communication module.

[0028] Furthermore, the communication module includes a mobile communication module, a Wifi module and a Bluetooth module. The communication module sends the alarm information to the cloud server for remote monitoring and data recording through any one or more of the mobile communication module, the Wifi module and the Bluetooth module.

[0029] Furthermore, the transmitting module includes a transmitting antenna and a transmitting unit module, and the transmitting antenna is provided with two groups both connected to the transmitting unit module;

[0030] The receiving module includes a receiving antenna and a receiving unit module, and the receiving antenna is provided with two groups both connected to the transmitting unit module;

[0031] The transmitting module transmits the signal specifically including the following process: the millimeter wave transmitting antenna generates a high-frequency millimeter wave signal through the transmitting unit module and transmits it into the detection area. When the transmitted millimeter wave signal encounters the target object, the millimeter wave signal will be scattered and reflected by the different structures and physical properties of the surface of the target object;

[0032] The receiving module receives the signal specifically including the following process: the receiving unit module is responsible for receiving the millimeter wave echo signal reflected from the target object, and according to the relative motion between the target object and the radar, the receiving unit module can receive the millimeter wave echo signal reflected from the target object.

[0033] Further, an electronic device includes a processor and a storage device;

[0034] The storage device stores a computer program, and the processor executes the computer program to perform the method described above.

[0035] Furthermore, a computer storage medium includes a computer program and data stored therein, wherein the computer program implements the above method when executed by the processor.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] 1. The present invention has good privacy protection, high detection sensitivity, high accuracy, and strong environmental adaptability. It can realize the functions of long-term continuous real-time human presence, falls, stationary residence, and recognition of voice help, reducing the situation where elderly people living alone cannot get timely help after falling. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 This is a system diagram of the off-line voice confirmation device for fall detection of the present invention;

[0039] Figure 2 This is a system diagram of a millimeter wave radar module in the offline voice confirmation device for fall detection of the present invention;

[0040] Figure 3 This is a system diagram of an offline voice interaction module in the offline voice confirmation device for fall detection of the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] A method for offline voice confirmation based on millimeter wave radar fall detection specifically comprises the following steps:

[0043] Step (a): millimeter wave radar monitoring; after being turned on, the millimeter wave radar module 1 continuously transmits millimeter wave signals and receives reflected signals. The microprocessor 3 processes the millimeter wave radar module 1 signal through the signal processing algorithm to obtain the target height information. The system performs real-time monitoring based on the calculated height information and compares the current height H current and dynamic height threshold H th , when H current <H th When the system initially determines that the target may fall;

[0044] Step (b): voice recognition confirmation; after the height information is lower than the set threshold, the system activates the microphone, starts to capture the audio signal of the surrounding environment, records the audio data, processes the audio signal through the voice recognition unit module 201 recognition chip, and recognizes specific voice commands. The system has a built-in voice library that can recognize common help phrases, such as "I fell down", "Please help", etc. If the help signal is successfully recognized, the system further confirms that the target is indeed in a falling state;

[0045] Step (c): Multiple confirmation mechanism: To improve accuracy, the system is designed with a multiple confirmation mechanism. When the height information is lower than the threshold for multiple consecutive times and the voice recognition results are consistent, the system confirms that the target is in a falling state.

[0046] Step (d): Upload information; once the target falls, the system uploads relevant information, such as timestamp, user height, fall confirmation status, recognized voice commands, etc., to the cloud server 5 through the communication module 4. The cloud server 5 can receive and store data in real time for subsequent analysis and monitoring. The user or his relatives can access the cloud server 5 at any time through a mobile device or computer to view the fall event record and related statistical data to achieve remote monitoring.

[0047] In order to improve the accuracy of fall detection, the system needs to dynamically adjust the threshold value according to the real-time height information. In step (a), the dynamic height threshold H th Fitting, the specific method is as follows:

[0048] Data collection: The system continuously monitors and records the target's altitude data, forming a height time series data set D = {H1, H2..., H n}, where H i Represents the height information at the i-th time point;

[0049] Linear regression model: Use a linear regression model to fit the height data to determine the dynamic height threshold. The model assumes: H fit =a·t+h, where H fit is the height of the fit, t is the time, a is the slope, and b is the intercept;

[0050] Least Squares Method: The model parameters a and b are calculated by the least squares method to minimize the following sum of squared errors:

[0051] Dynamic height threshold generation: After the model parameters are determined, the dynamic height threshold H is generated according to the height change of the target. th , can be set as a percentage of the fitting result, for example: H fit =H fit (1-δ) where δ is a constant less than 1 and is used to adjust the sensitivity of the threshold;

[0052] “Adaptive adjustment: The system can adaptively adjust the parameter δ according to environmental changes and individual characteristics to ensure that the dynamic altitude value can effectively reflect the normal altitude range of the target.

[0053] An offline voice confirmation device based on millimeter wave radar fall detection, comprising:

[0054] The microprocessor 3 is respectively connected to the millimeter wave radar module 1, the offline voice interaction module 2, and the communication module 4. The communication module 4 is connected to the cloud server 5. The cloud server 5 is used to receive the alarm information and help information of the offline voice interaction module 2 sent by the communication module 4.

[0055] Millimeter wave radar module 1; the millimeter wave radar module 1 is used to send, receive and process millimeter wave signals, and store preprocessing programs for processing millimeter wave signals;

[0056] The millimeter wave radar module 1 includes:

[0057] Transmitting module; the transmitting module is used to transmit millimeter wave signals; the transmitting module includes a transmitting antenna 101 and a transmitting unit module 103, and the transmitting antenna 101 is provided with two groups both connected to the transmitting unit module 103; the transmitting module transmits signals specifically including the following process: the millimeter wave transmitting antenna 101 generates a high-frequency millimeter wave signal through the transmitting unit module 103, and transmits it into the detection area, when the transmitted millimeter wave signal encounters a target object, the millimeter wave signal will be scattered and reflected by different structures and physical properties on the surface of the target object;

[0058] Receiving module; the receiving module is used to receive the millimeter wave echo signal emitted by the transmitting module and scattered and reflected by the target object; the receiving module includes a receiving antenna 102 and a receiving unit module 104, and the receiving antenna 102 is provided with two groups of transmitting unit modules 103; the receiving module receives the signal specifically including the following process: the receiving unit module 104 is responsible for receiving the millimeter wave echo signal reflected from the target object, and according to the relative movement of the target object and the radar, the receiving unit module 104 can receive the millimeter wave echo signal reflected by the target object.

[0059] Control unit module 105; the control unit module 105 is used to control the receiving module to receive the millimeter wave echo signal and pre-process the millimeter wave echo signal; the storage unit module 106, the storage unit module 106 is used to store the pre-processing program for processing the millimeter wave echo signal;

[0060] The control unit module 105 is connected to the transmitting module, the receiving module and the storage unit module 106 respectively.

[0061] Microprocessor 3; Microprocessor 3 is used to control the operation, calculation and storage of the module, process the millimeter wave signal, measure the parameters of the target object and judge the fall, and send the fall signal; Microprocessor 3 is used to control the operation, calculation and storage of millimeter wave radar module 1, offline voice interaction module 2 and communication module 4; Millimeter wave signal processing specifically includes the following process: through the received millimeter wave multi-channel echo signal and the transmitted mixed intermediate frequency signal, the intermediate frequency signal is processed by fast Fourier transform to generate a preliminary spectrum diagram;

[0062] The target object parameter measurement specifically includes the following processes: by analyzing the changes in various parameters such as the intensity frequency and phase of the received millimeter wave signal, some important parameters of the target object, such as distance, speed, azimuth, etc., can be obtained. The distance, speed, azimuth and other parameters of the target object can reflect the three-dimensional position information and height data of the target object in the detection area; the target object fall judgment specifically includes the following processes: calculating the current height information obtained and the fitted dynamic height threshold, and using two values ​​(previous height H current and dynamic height threshold H th) to determine whether the activity target has fallen.

[0063] Offline voice interaction module 2; the offline voice interaction module 2 is used to receive the fall signal sent by the microprocessor 3, interact with the user, and convert it into warning information and help information according to the user feedback information. The offline voice interaction module 2 includes a voice recognition unit module 201, a microphone 202 and a speaker 203, and the voice recognition unit module 201, the microphone 202 and the speaker 203 are connected to each other; the conversion into warning information specifically includes the following process: the microprocessor 3 sends the fall signal to the offline voice interaction module 2, and the device broadcasts the fall confirmation message and the user feedback message through the speaker 203 and the microphone 202 device. Within the specified time, according to the user's shouting response feedback of whether the voice signal falls, it is converted into a fall or release alarm message command after recognition by the voice recognition unit module 201, and then the corresponding alarm message is sent to the cloud server 5 through the communication module 4; the conversion into help information specifically includes the following process: when the user feels unwell, he can actively call for help, the device microphone 202 device collects the call voice signal, and the rescue command is recognized by the voice recognition chip in the offline voice interaction module 2, and the microprocessor 3 sends the rescue command to the cloud server 5 through the communication module 4.

[0064] Communication module 4, communication module 4 is used to receive the alarm information and help information commands of the offline voice interaction module 2, and send the alarm information and help information; the communication module 4 includes a mobile communication module, a Wifi module and a Bluetooth module, and through any one or more of the mobile communication module, the Wifi module and the Bluetooth module, the alarm information is sent to the cloud server 5 for remote monitoring and data recording.

[0065] An electronic device includes a processor and a storage device; the storage device stores a computer program, and the processor executes the computer program to perform the above method.

[0066] A computer storage medium includes a computer program and data stored therein, and the computer program implements the above method when executed by a processor.

[0067] The present invention has good privacy protection, high detection sensitivity, high accuracy, and strong environmental adaptability. It can realize the functions of long-term continuous real-time human presence, falls, stationary residence, and recognition of voice help, reducing the situation where elderly people living alone cannot get timely help after falling.

[0068] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for offline voice confirmation based on millimeter wave radar fall detection, characterized in that: The specific steps include: Step (a): Millimeter-wave radar monitoring: After being turned on, the millimeter-wave radar module continuously transmits millimeter-wave signals and receives reflected signals. The millimeter-wave radar module signal is processed by the microprocessor signal processing algorithm to obtain the target's height information. The system performs real-time monitoring based on the calculated height information and compares the current height H current and dynamic height threshold H th , when H current <H th When the system initially determines that the target may fall; Step (b): voice recognition confirmation; after the height information is lower than the set threshold, the system activates the microphone, starts to capture the audio signal of the surrounding environment, records the audio data, processes the audio signal through the offline voice interaction module recognition chip, and recognizes specific voice commands. The system has a built-in voice library that can recognize common help phrases, such as "I fell down" and "Please help", etc. If the help signal is successfully recognized, the system further confirms that the target is indeed in a falling state; Step (c): Multiple confirmation mechanism: To improve accuracy, the system is designed with a multiple confirmation mechanism. When the height information is lower than the threshold for multiple consecutive times and the voice recognition results are consistent, the system confirms that the target is in a falling state. Step (d): Upload information; once the target falls, the system uploads relevant information, such as timestamp, user height, fall confirmation status, recognized voice commands, etc., to the cloud server through the communication module. The cloud server can receive and store data in real time for subsequent analysis and monitoring. The user or his relatives can access the cloud server at any time through a mobile device or computer to view the fall event record and related statistical data, thereby realizing remote monitoring.

2. An offline voice confirmation device for fall detection based on millimeter wave radar according to claim 1, characterized in that: include: Millimeter wave radar module; The millimeter wave radar module is used to send, receive and process millimeter wave signals, and store preprocessing programs for processing millimeter wave signals; microprocessor; The microprocessor is used to control the operation, calculation and storage of the module, process the millimeter wave signal, measure the parameters of the target object and determine the fall, and send the fall signal; Offline voice interaction module; the offline voice interaction module is used to receive the fall signal sent by the microprocessor, interact with the user, and convert the user feedback information into warning information and help information; A communication module, the communication module is used to receive the alarm information and help information commands of the offline voice interaction module, and send the alarm information and help information; The microprocessor is respectively connected to the millimeter wave radar module, the offline voice interaction module, and the communication module. The communication module is connected to the cloud server. The cloud server is used to receive the alarm information and help information of the offline voice interaction module sent by the communication module.

3. The offline voice confirmation device based on millimeter wave radar fall detection according to claim 2 is characterized in that: The millimeter wave radar module includes: Transmitting module; the transmitting module is used to transmit millimeter wave signals; Receiving module; the receiving module is used to receive the millimeter wave echo signal emitted by the transmitting module and scattered and reflected by the target object; Control unit module; the control unit module is used to control the receiving module to receive the millimeter wave echo signal and pre-process the millimeter wave echo signal; A storage unit module, the storage unit module is used to store a preprocessing program for processing millimeter wave echo signals; Wherein, the control unit module is respectively connected to the transmitting module, the receiving module and the storage unit module.

4. The offline voice confirmation device based on millimeter wave radar fall detection according to claim 2 is characterized in that: The microprocessor is used to control the operation, calculation and storage of the millimeter wave radar module, the offline voice interaction module and the communication module.

5. The offline voice confirmation device based on millimeter wave radar fall detection according to claim 2 is characterized in that: The millimeter wave signal processing specifically includes the following process: by processing the received millimeter wave multi-channel echo signal and the transmitted mixed intermediate frequency signal, using fast Fourier transform to process the intermediate frequency signal, and generate a preliminary spectrum diagram; The target object parameter measurement specifically includes the following process: by analyzing the changes in various parameters such as the intensity frequency and phase of the received millimeter wave signal, some important parameters of the target object, such as distance, speed, azimuth, etc., can be obtained. The parameters can reflect the three-dimensional position information and height data of the target object in the detection area; The target object fall judgment specifically includes the following process: calculating the acquired current height information and the fitted dynamic height threshold, and judging whether the moving target falls by comparing the two values.

6. The offline voice confirmation device based on millimeter wave radar fall detection according to claim 2, characterized in that: The conversion into alarm information specifically includes the following process: the microprocessor sends the fall signal to the offline voice interaction module, the device broadcasts the fall confirmation message and the user feedback message through the speaker and microphone device, and within the specified time, the voice signal of whether the user shouts and responds to the feedback of the fall is converted into a fall or alarm release message command after being recognized by the voice recognition system, and then the corresponding alarm message is sent to the cloud server through the communication module; The conversion into help information is specific The process includes the following: when the user feels unwell, he can actively call for help. The device microphone collects the call voice signal, and the voice recognition chip in the offline voice interaction module recognizes the rescue command. The microprocessor then sends the rescue command to the cloud server through the communication module.

7. The device for offline voice confirmation based on millimeter wave radar fall detection according to claim 2, characterized in that: The communication module includes a mobile communication module, a Wifi module and a Bluetooth module. The communication module sends the alarm information to the cloud server for remote monitoring and data recording through any one or more of the mobile communication module, the Wifi module and the Bluetooth module.

8. The device for offline voice confirmation based on millimeter wave radar fall detection according to claim 2, characterized in that: The transmitting module includes a transmitting antenna and a transmitting unit module, and the transmitting antenna is provided with two groups both connected to the transmitting unit module; The receiving module includes a receiving antenna and a receiving unit module, and the receiving antenna is provided with two groups both connected to the transmitting unit module; The transmitting module transmits the signal specifically including the following process: the millimeter wave transmitting antenna generates a high-frequency millimeter wave signal through the transmitting unit module and transmits it into the detection area. When the transmitted millimeter wave signal encounters the target object, the millimeter wave signal will be scattered and reflected by the different structures and physical properties of the surface of the target object; The receiving module receives the signal specifically including the following process: the receiving unit module is responsible for receiving the millimeter wave echo signal reflected from the target object, and according to the relative motion between the target object and the radar, the receiving unit module can receive the millimeter wave echo signal reflected from the target object.

9. An electronic device, characterized in that: including a processor and a storage device; The storage device stores a computer program, and the processor executes the computer program to perform the method as claimed in claim 1.

10. A computer storage medium, comprising a computer program and data stored therein, wherein the computer program implements the method as claimed in claim 1 when executed by the processor.

Images