Fall detection method and device for intelligent pedestal pan and intelligent pedestal pan

Abstract

The invention relates to the technical field of intelligent pedestal pans, in particular to a falling detection method and device for an intelligent pedestal pan and the intelligent pedestal pan. The fall detection method comprises the following steps: an information acquisition step; the height and the phase angle of a target person are detected in real time through a microwave radar located on the intelligent pedestal pan; whether a person sits or not is detected through a sitting sensor on the intelligent pedestal pan; a fall judgment step; and after the sitting sensor detects that the seat ring is changed from a seated state to a non-seated state, if the height of the target person monitored by the microwave radar is below a preset height, the duration time of the state reaches a preset time or above, and the target person stays in the interval of a preset phase angle, judging that the target person is in a falling state. Through the steps, the accuracy and reliability of detection can be effectively improved, and the false alarm rate is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent toilets, and particularly to a fall detection method, device and intelligent toilet for intelligent toilets. Background Art

[0002] In modern society, with the deepening of the aging degree, the health and safety of the elderly have received increasing attention. The bathroom is a place where accidents occur frequently in the family, and fall incidents occur frequently, seriously threatening the life and health of users. According to statistics, falls have become the leading cause of injury-related deaths among the elderly, and falls that occur during the use of the toilet in the bathroom are often difficult to be discovered by others in time due to the relatively hidden location, resulting in the inability of the fallen person to obtain assistance in the first time, delaying the best treatment opportunity and causing serious consequences.

[0003] At present, the technical means for fall detection during the use of the toilet in the bathroom on the market are relatively limited and have obvious deficiencies. Some detection systems rely on cameras for image recognition, which not only have high costs but also involve the problem of user privacy protection and are difficult to be popularized and used in private spaces. And some sensor-based detection methods have the defects of low detection accuracy and high false alarm rate, and cannot accurately identify the fall behavior of users. For fall detection in the toilet use scenario, the existing detection schemes can neither cover the surrounding area of the toilet comprehensively nor make accurate judgments by integrating multiple pieces of information, resulting in the inability to detect fall incidents in time and reliably in actual applications and being difficult to meet the urgent needs of the toilet safety function in the intelligent health care field.

[0004] Therefore, there is an urgent need for a technical solution that can detect the fall behavior during the use of the toilet in real time and accurately to fill the market gap and provide a safer and more reliable health care guarantee for users. Summary of the Invention

[0005] To solve the deficiencies in detecting the fall behavior during the use of the toilet in the bathroom in the above-mentioned prior art, the present invention provides a fall detection method for an intelligent toilet, including the following steps: Information acquisition step; the height of the target person and the phase angle of the target person are detected in real time by a microwave radar located on the intelligent toilet; whether there is someone sitting is detected by a seat sensor on the intelligent toilet. Fall determination step; after the seat sensor detects that the seat ring changes from being occupied by someone to being unoccupied, if the height of the target person monitored by the microwave radar is below a preset height, and this state lasts for a preset time or more, and at the same time the target person stays within the interval of the preset phase angle, it is determined that the target person is in a fallen state.

[0006] The present invention also provides a fall detection device for an intelligent toilet, including: An information acquisition module; configured to detect the height of a target person and the phase angle of the target person in real time through a microwave radar located on a smart toilet; and detect whether there is someone sitting on the seat ring through a seat sensor on the smart toilet. A fall determination module; after the seat sensor detects that the seat ring changes from being occupied to unoccupied, if the height of the target person monitored by the microwave radar is below a preset height, and this state lasts for a preset time or more, and at the same time the target person stays within a preset phase angle range, it is determined that the target person is in a fallen state.

[0007] The present invention also provides a smart toilet, which adopts the fall detection method for a smart toilet described in the above embodiment or the fall detection device for a smart toilet described in the above embodiment.

[0008] Based on the above, compared with the prior art, the fall detection method for a smart toilet provided by the present invention can effectively determine whether a target person is in a fallen state through signal detection and determination processing of applying a microwave radar to a smart toilet. Its technology is simple, efficient, and safe, without the need for complex computing power, and can effectively improve the accuracy and reliability of detection and reduce the false alarm rate.

[0009] Other features and beneficial effects of the present invention will be described in the subsequent specification, and some of them will become obvious from the specification or be understood by implementing the present invention. The objectives and other beneficial effects of the present invention can be achieved and obtained through the structures specifically pointed out in the specification, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for describing the embodiments or the prior art. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings; in the following description, the positional relationships in the drawings, unless otherwise specified, are based on the directions in which the components are shown in the drawings.

[0011] Figure 1 It is a block diagram of the steps of the fall detection method for a smart toilet provided by an embodiment of the present invention. Figure 2 It is a flowchart of the steps of the fall detection method for a smart toilet provided by an embodiment of the present invention. Figure 3 、 Figure 4 It is a block diagram of the steps of the fall detection method for a smart toilet provided by different embodiments of the present invention. Figure 5, Figure 6 The structural block diagram of the fall detection device for a smart toilet provided by different embodiments of the present invention. Detailed implementation manners

[0012] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some but not all of the embodiments of the present invention. The technical features designed in different implementation manners of the present invention described below can be combined with each other as long as they do not conflict with each other. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0013] In the description of the present invention, it should be noted that all terms used in the present invention (including technical terms and scientific terms) have the same meanings as those commonly understood by those of ordinary skill in the art to which the present invention belongs, and should not be construed as limiting the present invention. It should be further understood that the terms used in the present invention should be understood as having meanings consistent with their meanings in the context of this specification and the relevant art, and should not be understood in an idealized or overly formal sense, unless otherwise clearly defined in the present invention.

[0014] Embodiment 1 Please refer to Figure 1 , the present invention provides a fall detection method for a smart toilet, including the following steps: Information acquisition step: The height of the target person and the phase angle of the target person are detected in real time by a microwave radar located on the smart toilet; whether there is someone sitting is detected by a seat sensor on the smart toilet. Fall determination step: After the seat sensor detects that the seat ring changes from being occupied by someone to being unoccupied, if the height of the target person monitored by the microwave radar is below a preset height, and this state lasts for a preset time or more, and at the same time the target person stays within a preset phase angle range, it is determined that the target person is in a fallen state.

[0015] Through the above detection method, the position of the human body in the bathroom space can be tracked in real time, effectively overcoming the interference of the complex bathroom environment on the microwave radar positioning, ensuring that the system can accurately reflect the dynamic changes of the human body, and improving the accuracy and reliability of the human body position detection.

[0016] In the information acquisition step, a microwave radar and a seating sensor are installed on the intelligent toilet. The microwave radar needs to be installed at a position that can comprehensively cover a certain range of space around the toilet to ensure that the height and phase angle of the target person can be detected in real time. For example, in this embodiment, it is preferred that the millimeter-wave radar is placed in the electronic control box of the intelligent toilet and detects the height of the target person and the phase angle of the target person at a detection range angle of 120 degrees; the electronic control box is located behind the base of the intelligent toilet and is more than a preset height from the ground to accurately scan a detection range of 120°, and at the same time can effectively cover the area where a person may fall. The seating sensor is installed at a suitable position on the seat ring to accurately detect whether there is someone sitting on the seat ring.

[0017] At the same time, key parameters are preset in advance, including but not limited to a preset height, a preset time, and a preset phase angle. Among them, the preset height is set according to the height difference between the normal standing and falling states of the human body, and the value of the preset height ranges between 40 and 60 cm; for example, it is set to 50 cm or 60 cm. The preset time can generally be set to more than 10 seconds to be used to judge the duration of the target person in the low-height state; in this embodiment, it is preferred that the value of the preset time ranges between 20 and 40 s, for example, 30 s; the range of the preset phase angle is determined according to the normal activity range around the intelligent toilet and the possible positions where a fall may occur. For example, it is set to a phase angle range of 30° - 60° directly in front of the toilet. It is specifically set reasonably according to actual needs.

[0018] Please refer to Figure 2 , when specifically acquiring information, the microwave radar scans the space around the intelligent toilet at a certain time interval (such as every 0.1 second) to obtain the height and phase angle data of the target person in real time. The seating sensor senses the information of whether there is someone sitting by skin contact or continuously monitoring the pressure change on the seat ring. To reduce energy consumption, when the microwave radar detects the dynamic activity of the target person, the acquisition of the height and phase angle data of the target person is started.

[0019] Please continue to refer to Figure 3, in the fall determination step, after receiving the information sent by the seat sensor that the seat ring has changed from being occupied to unoccupied, first perform height judgment to determine whether the height of the target person detected by the microwave radar is below the preset height. If the height of the target person is higher than the preset height, it is determined that the target person is not in a fallen state, and information acquisition and monitoring continue. If the height of the target person is below the preset height, start timing and record the duration of this low-height state. Then perform time determination. When the duration of this state reaches or exceeds the preset time, enter the next phase angle judgment to determine whether the target person stays within the interval of the preset phase angle. If the target person is within this interval, it is determined that the target person is in a fallen state; if the target person is not within this interval, it is determined that the target person is not in a fallen state, and information acquisition and monitoring continue. It should be noted that this embodiment is not limited to the above examples and Figure 2 the step judgment order shown. For example, in other methods, the phase angle judgment can be performed first, then the height judgment, and finally the time judgment, etc.

[0020] Furthermore, in order to effectively avoid the actions of the target person at other positions from interfering with the detection results. This embodiment also preferably includes, in the fall determination step, a determination of whether the target person is within the preset distance range. That is, when the seat sensor detects that the seat ring has changed from being occupied to unoccupied, if the height of the target person detected by the microwave radar is below the preset height, and the duration of this state reaches or exceeds the preset time, and at the same time the target person stays within the interval of the preset phase angle, and the target person is within the preset distance range, it is determined that the target person is in a fallen state. Among them, the preset distance range can be adjusted according to the actual space of the bathroom. For example, in this embodiment, the preset distance range is preferably within 2 meters in front of the intelligent toilet that the radar can detect. This judgment design can better adapt to the bathroom space layouts in various different scenarios, prevent interference, and improve the application range at the same time.

[0021] Through the above steps of determination, the intelligent toilet can effectively detect the fall situation of the target person. Its technology is simple, efficient, and safe, without the need for complex computing power, and can effectively improve the accuracy and reliability of detection and reduce the false alarm rate.

[0022] Furthermore, when it is determined that the target person is in a fallen state, the control system of the intelligent toilet triggers an alarm mechanism. A loud alarm sound can be emitted through the built-in speaker, and at the same time, the fall information is sent to the pre-bound mobile phone APP or other intelligent devices through a wireless communication module (such as WiFi or Bluetooth) to notify relevant personnel.

[0023] Alternatively, when it is determined that the target person is in a fallen state, the system can further confirm the status of the target person by asking "Do you need help?" through intelligent voice; if the response is yes or there is no response, an emergency notification will be pushed to family members through the mobile phone APP or the smart home central control; in severe cases, the system can directly connect to the community medical service or the emergency rescue center, and retain or transmit key information such as the fall time and continuous status. Of course, the system can also include data backup and analysis steps to record the behavioral characteristics before the fall, help family members or medical staff evaluate the cause of the fall, provide data support for subsequent preventive measures, and form a complete closed-loop for the safety guarantee of the elderly in the toilet.

[0024] In an alternative embodiment, please refer to Figure 3 , and it further includes a first information processing step; by obtaining the height values of the target person continuously collected at fixed time intervals, a real-time height measurement value sequence is obtained; for the real-time height measurement value sequence, the difference between adjacent height values is calculated, and the difference is normalized at fixed time intervals to generate a data sequence of height change rates; the first derivative data is extracted from the data sequence of height change rates, and the numerical differentiation method is used to calculate the change amount between adjacent rates to obtain a derivative sequence reflecting the characteristics of the action speed; based on the analysis of the derivative sequence, the rate change trend of the target person's action is obtained, and it is judged whether there is a sudden rate anomaly point.

[0025] When determining whether there is a sudden rate anomaly point, it is determined based on an anomaly threshold. If the detected rate change is greater than the anomaly threshold, it is determined that there is a sudden rate anomaly point.

[0026] Specifically, when implemented, the microwave radar continuously collects the height values of the target person at fixed time intervals (for example ). Assume that the height values collected at time points are respectively , thus obtaining a real-time height measurement value sequence ; for the real-time height measurement value sequence , calculate the difference between adjacent height values , and the formula is: ; where ; through the fixed time interval normalize the difference to obtain the height change rate , and the formula is: ; where ; that is, a data sequence of height change rates is formed; from the data sequence of height change rates Extract the first derivative data. The numerical differentiation method is used to calculate the change between adjacent rates, and a derivative sequence reflecting the characteristics of the action speed is obtained. ; The derivative The calculation formula is:

[0027] In the formula, ; Based on the derivative sequence Analyze the rate change trend of the target person's actions. Set a high anomaly threshold , when the absolute value of the derivative is greater than this high anomaly threshold , that is , then determine that this point is a sudden rate anomaly point.

[0028] In the fall determination step, if there is a sudden rate anomaly point, after the seat sensor detects that the seat ring changes from being occupied to unoccupied, when the microwave radar monitors that the height of the target person is below the preset height and lasts for more than the preset time, and at the same time the target person stays in the interval of the preset phase angle, it is determined that the target person is in a falling state, so as to more accurately and reliably detect the falling state of the target person and further reduce the false alarm rate.

[0029] In another alternative embodiment, please refer to Figure 3 , it also includes a second information processing step, which filters the obtained phase angle of the target person and extracts features to calculate the angular velocity change rate of the phase angle; based on the angular velocity change rate of the phase angle, it is determined whether there is a sudden angular velocity change anomaly point.

[0030] Specifically, the microwave radar continuously collects the phase angles of the target person at a fixed time interval (for example ), and the obtained phase angle sequence is , getting the phase angle sequence ; Use a digital filter (such as a low-pass filter) or a common filtering algorithm to filter the obtained phase angle sequence to remove noise interference. Taking simple moving average filtering as an example, the filtered phase angle The calculation formula is: (when ), where is the filter window size and can be adjusted according to the actual situation. For cases, the average value of the first phase angles can be used for filtering. Calculate the difference between adjacent filtered phase angles as , ; By a fixed time interval normalize the difference to obtain the angular velocity of the phase angle which is: , ; Then calculate the change amount between adjacent angular velocities to obtain the angular velocity change rate of the phase angle which is:

[0031] In the formula, .

[0032] Set an abnormal threshold for angular velocity change , when the absolute value of the angular velocity change rate is greater than this abnormal threshold for angular velocity change , that is , then determine that this point is a sudden abnormal point of angular velocity change.

[0033] In the fall determination step, if there is a sudden abnormal point of angular velocity change, after the seat sensor detects that the seat ring changes from being occupied to unoccupied, when the microwave radar monitors that the height of the target person is below the preset height and lasts for more than the preset time, and at the same time the target person stays in the interval of the preset phase angle, it is determined that the target person is in a falling state, so as to more accurately and reliably detect the falling state of the target person and further reduce the false alarm rate.

[0034] In other alternative embodiments, please refer to Figure 3 , in the information acquisition step, it further includes obtaining the horizontal width of the target person in real time based on the microwave radar.

[0035] It further includes a third information processing step. By obtaining the horizontal width values of the target person collected continuously at fixed time intervals, a real-time width measurement value sequence is obtained; for the real-time width measurement value sequence, calculate the difference between adjacent horizontal width values, and normalize the difference at fixed time intervals to generate a data sequence of horizontal width change rates; based on the horizontal width change rate, determine whether there is a sudden abnormal point of horizontal width change.

[0036] Specifically, the microwave radar continuously collects the horizontal width values of the target person at a fixed time interval (for example ). Assume that the horizontal width values collected at time points are respectively , thus obtaining a real-time width value sequence ; calculate the difference between adjacent width values , and the formula is: ; where ; at fixed time intervals normalize the difference to obtain the horizontal width change rate , and the formula is: ; where ; that is, a data sequence of the horizontal width change rate is formed ; set an abnormal angular velocity change threshold , when the absolute value of the angular velocity change rate is greater than this abnormal angular velocity change threshold , that is , then determine that this point is a sudden abnormal point of horizontal width change.

[0037] In the fall determination step, if there is a sudden abnormal point of horizontal width change, after the seat sensor detects that someone has taken a seat and then the seat is empty, when the microwave radar monitors that the height of the target person is below the preset height and lasts for more than the preset time, and at the same time the target person stays in the interval of the preset phase angle, it is determined that the target person is in a fallen state, so as to more accurately and reliably detect the fallen state of the target person and further reduce the false alarm rate.

[0038] Preferably, in order to further improve the accuracy of fall detection, please refer to Figure 4 , this embodiment further provides a method for capturing a special "motion ripple" generated at the moment of a fall based on Doppler micro-motion characteristics to assist in determining whether there is a fall behavior. Specifically, it further includes a fourth information processing step for extracting and processing the Doppler micro-motion characteristic data in the signal of the microwave radar, and constructing a time-frequency analysis model of the Doppler frequency shift changing with time, and identifying the special "motion ripple" generated at the moment of a fall from it; the special "motion ripple" is manifested as a characteristic pattern with a specific frequency range, duration and energy distribution that appears in the time-frequency analysis model, and this pattern is different from the time-frequency characteristics when the target person is not at the moment of a fall.

[0039] Specifically, in the fourth information processing step, the microwave radar signal can be preprocessed first, including operations such as filtering and denoising to improve the signal quality. Then, signal processing algorithms (such as short-time Fourier transform, wavelet transform, etc.) are used to extract Doppler micro-motion feature data from the preprocessed signal. These data reflect the Doppler frequency shift changes generated by various parts of the target person's body during micro-motions. Next, based on the extracted Doppler micro-motion feature data, a time-frequency analysis model of the Doppler frequency shift changing with time is constructed. This model can intuitively display the distribution of the Doppler frequency shift at different times and frequencies. For example, a time-frequency diagram (such as a spectrogram) is used to represent the time-frequency analysis model, with the abscissa representing time and the ordinate representing frequency. Finally, in the constructed time-frequency analysis model, through preset algorithms and rules, the special "motion ripples" generated at the moment of a fall are identified. The special "motion ripples" are characterized by a feature pattern that appears in the time-frequency analysis model with a specific frequency range (e.g., 5 - 20 Hz), duration (e.g., 0.5 - 2 seconds), and energy distribution. This pattern is significantly different from the time-frequency characteristics of the target person when not in the moment of falling (such as normal standing, sitting, walking, etc.). To accurately identify this feature pattern, a large amount of time-frequency data of the target person in different states can be collected in advance to establish a feature template library, and the real-time data is compared with the features in the template library for identification.

[0040] Based on the above processing of the fourth information step, in the fall determination step, after the seat sensor detects that the seat ring changes from being occupied to unoccupied, if the height of the target person monitored by the microwave radar is below the preset height and this state lasts for more than the preset time, and at the same time the target person stays within the preset phase angle range, and a signal conforming to the feature pattern of the special "motion ripples" is detected in the time-frequency analysis model, it is determined that the target person is in a fallen state.

[0041] Through the above specific implementation, by extracting the Doppler micro-motion feature data and capturing the special "action ripples" in the fallen state, the accuracy of fall determination of the intelligent toilet can be effectively improved, providing more reliable safety protection for users.

[0042] In this embodiment, to improve the accuracy of fall determination, as Figure 4 shown, the ultra-high resolution of the millimeter-wave radar can also be used to detect the displacement characteristics of internal organs and / or muscle tremor characteristics to assist in determining the fall behavior.

[0043] Among them, the method for assisting in determining the fall behavior based on the displacement characteristics of internal organs includes: In the information acquisition step, it further includes real-time collection of the displacement data of the internal organs of the target person based on a microwave radar; it further includes a fifth information processing step of pre-establishing an organ movement baseline database, which contains the movement characteristic data of the internal organs of the target person in a normal state; comparing and analyzing the displacement data of the internal organs with the data in the organ movement baseline database, and confirming whether there are mutation differences between the displacement data of the internal organs and the data in the organ movement baseline database and whether the mutation differences conform to the characteristics of inertial micro-displacement of internal organs during a fall. Specifically, using the high-resolution characteristics of the microwave radar, the displacement data of the internal organs of the target person is collected in real time. For example, a 60GHz radar can be used to detect a 0.2mm-level heart displacement. Among them, the microwave signal emitted by the microwave radar can penetrate human tissues. When the internal organs move, the characteristics of the reflected signal will change accordingly. By analyzing and processing these reflected signals, the displacement data of the internal organs can be obtained. In the fifth information processing step, the pre-established organ movement baseline database contains the movement data of the internal organs of the target person continuously collected by the microwave radar in the normal state of the target person (such as non-falling states like standing and walking). By analyzing and processing the collected data, characteristic parameters of the movement of the internal organs are extracted, such as the displacement range, movement frequency, movement speed, etc. These characteristic parameters are stored in the storage unit to form an organ movement baseline database. Based on the organ movement baseline database, the fifth information processing step will compare and analyze the displacement data of the internal organs collected in real time with the data in the organ movement baseline database. Specifically, the difference values of the two in characteristic parameters such as displacement range, movement frequency, and movement speed can be calculated. It is judged whether there are mutation situations in these difference values, that is, whether there are obvious changes beyond the normal fluctuation range in a short time. At the same time, it is checked whether these mutation differences conform to the characteristics of inertial micro-displacement of internal organs during a fall. For example, during a fall, the organs in the abdominal cavity may produce an instantaneous displacement forward or to one side due to inertia, and its displacement speed and amplitude will be significantly different from the normal state. Through the above judgment, it is determined whether there are mutation differences in the displacement data of the internal organs that conform to the characteristics of inertial micro-displacement of internal organs during a fall.

[0044] Based on the determination of the displacement data of the internal organs in the above-mentioned fifth information processing step, in the fall determination step, when the seat sensor detects that the seat ring changes from being occupied to unoccupied, if the height of the target person monitored by the microwave radar is below the preset height, and this state lasts for more than the preset time, and the target person stays within the interval of the preset phase angle, and there are mutation differences in the displacement data of the internal organs that conform to the characteristics of inertial micro-displacement of internal organs during a fall, it is determined that the target person is in a fallen state.

[0045] Among them, the method for assisting in determining a fall behavior based on the muscle tremor characteristics includes: In the information acquisition step, it further includes real-time collection of tremor signals generated by the muscle activities of the target person based on a microwave radar; it further includes a sixth information processing step for processing and analyzing the tremor signals to generate a muscle tremor map; comparing the generated muscle tremor map with the characteristic tremor signals in the preset range of 9 - 13 Hz to determine whether there are characteristic tremor signals in the muscle tremor map that conform to the range of 9 - 13 Hz.

[0046] Specifically, this embodiment also assists in determining the fall behavior based on the principle of 9 - 13 Hz characteristic tremors generated by stress-induced muscle contractions in the event of a fall. First, a microwave radar is used to collect the tremor signals generated by the muscle activities of the target person in real time. Since the minute displacements caused by the muscle tremors of the target person will cause corresponding changes in the reflected microwave signals, the tremor signals generated by the muscle activities can be obtained through precise detection and collection of these reflected signals. In the sixth information processing step, it can be preprocessed first, including filtering operations to remove environmental noise and other interference signals to improve the signal quality. Then, a suitable signal analysis method (such as Fourier transform) is used to convert the tremor signals in the time domain to the frequency domain and analyze the frequency components of the signals. On this basis, a muscle tremor map is generated, which intuitively shows the energy distribution of the tremor signals at different frequencies. Then, the generated muscle tremor map is compared with the characteristic tremor signals in the preset range of 9 - 13 Hz stored in the storage module. During the specific comparison process, the energy distribution characteristics of the muscle tremor map in the frequency range of 9 - 13 Hz are analyzed, such as parameters such as the peak intensity and frequency bandwidth in this frequency range, and compared with the corresponding parameters of the preset characteristic tremor signals to determine whether there are characteristic tremor signals in the muscle tremor map that conform to the range of 9 - 13 Hz.

[0047] Based on the determination of the above muscle tremor map, in the fall determination step, when the seat sensor detects that the seat ring changes from being occupied to unoccupied, if the height of the target person monitored by the microwave radar is below the preset height, and this state lasts for more than the preset time, and the target person stays within the interval of the preset phase angle, and there are characteristic tremor signals in the muscle tremor map that conform to the range of 9 - 13 Hz, it is determined that the target person is in a fall state.

[0048] Based on the above inventive concept, this embodiment can also use blood flow disturbance changes to assist in determining a fall behavior. Specifically, when a target person experiences sudden weightlessness (such as at the moment of a fall), significant changes occur in hemodynamics, which can cause changes in the velocity and direction of blood flow within blood vessels and small displacements of the blood vessel walls. These changes can be accurately detected by a microwave radar. It requires the microwave radar to have a displacement detection accuracy of 0.1 mm. Therefore, in the information acquisition step, a 0.1-mm microwave radar can be used to detect the blood flow disturbance data of the target person, and then a blood flow disturbance pattern database in the normal state and the fall state can be established. During the detection process, the blood flow disturbance data collected by the microwave radar in real time is compared and analyzed with the data in the database. If the real-time data matches the blood flow disturbance pattern in the fall state, that is, it shows a sudden change in blood flow velocity and blood vessel wall displacement characteristics significantly different from the normal state, and conforms to the characteristics of hemodynamic changes caused by sudden weightlessness, it can be used as one of the conditions for fall determination to improve the determination accuracy.

[0049] Optionally, during the fall behavior analysis process, it can also be used to assist in determining whether the target person has a fall behavior by detecting whether the fall caused by the target person getting up from the intelligent toilet has special characteristic behaviors.

[0050] For example, after the target person gets up, the height will first increase, but due to reasons such as unstable center of gravity, a fall occurs, so that although the height will increase, the expected maximum height range is not finally reached. Based on this special characteristic behavior, the following method can be used for auxiliary determination: In the fall determination step, when the seat sensor detects that the seat ring changes from being occupied by someone to unoccupied, if the height of the target person does not reach the detected maximum height range of the target person during the rising process, the height of the target person monitored by the microwave radar is below the preset height, and this state lasts for a preset time or more, and the target person stays within the preset phase angle range, it is determined that the target person is in a fall state. Among them, the setting of the maximum height range can be the proportional range of the maximum height that can be detected when the target person normally enters the detection area of the microwave radar. For example, the value range of the maximum height range can be between 4 / 5 times the maximum height and the maximum height, and it can be reasonably adjusted according to actual needs.

[0051] Furthermore, to avoid the influence of the moving cover or seat ring of the intelligent toilet on the microwave radar, in this embodiment, it is preferably set that the vertical area range where the intelligent toilet is located is an interference area. When it is detected that there is a microwave radar signal in the interference area and the seat ring and / or the cover rotates, the sensitivity of the microwave radar is reduced. Among them, sensors such as, but not limited to, angle sensors and Hall sensors can be used to monitor the rotation of the seat ring in real time. When both conditions of detecting a microwave radar signal in the interference area and the rotation of the seat ring and / or the cover are met, the sensitivity adjustment mechanism is triggered to reduce the sensitivity of the microwave radar. When the seat ring and / or the cover stops rotating, the microwave radar returns to the normal sensitivity level.

[0052] During the period when the sensitivity is reduced, the system will still continue to perform fall detection. However, due to the reduced sensitivity of the microwave radar, weak activity signals of the seat ring and / or the cover cannot be detected. Only signals that feedback large movements (such as falling) of the target person are detected. Therefore, the interference of the activities of the seat ring and / or the cover on the fall behavior detection can be effectively avoided. When the sensitivity of the microwave radar returns to normal, the system works according to the normal fall detection process to ensure the accurate determination of the fall state of the target person.

[0053] Through the above specific implementation manners, the influence of the moving cover or seat ring of the intelligent toilet on the microwave radar can be effectively avoided, and the detection accuracy and stability of the system can be improved.

[0054] More preferably, a machine learning algorithm can also be adopted. Specifically, by collecting the height and phase angle data of the target person detected by the microwave radar when different groups of people (the elderly, children, people with limited mobility, etc.) are using the toilet normally and performing various actions (bending down to pick up things, cleaning the toilet, squatting down to tie shoelaces, etc.), and at the same time marking the true states (normal, fall, etc.) corresponding to these data, pre-training is carried out based on these data through supervised learning algorithms such as decision trees, support vector machines, neural networks, etc. During the training process, the model parameters are continuously adjusted to enable the model to accurately distinguish different states. Then, the trained model is applied to the subsequent fall determination step. For example, when the model detects that the target person is in a fall state, the seat sensor detects that the seat ring changes from being occupied to unoccupied, and the height of the target person monitored by the microwave radar is below the preset height, and this state lasts for more than the preset time, and the target person stays in the interval of the preset phase angle, it is finally determined that the target person is in a fall state, thereby improving the accuracy of subsequent judgments.

[0055] It should be noted that in the fall determination step, such as Figure 3 , Figure 4As shown, after the seat sensor detects that the seat ring changes from being occupied to unoccupied, if the height of the target person monitored by the microwave radar is below the preset height, and this state lasts for more than the preset time, and at the same time the target person stays within the preset phase angle range, it can be combined with at least one of the following in the first information processing step, the second information processing step, the third information processing step, the fourth information processing step, the fifth information processing step, and the sixth information processing step: a sudden rate anomaly point, a sudden angular velocity change anomaly point, a sudden horizontal width change anomaly point, a signal conforming to the special "motion ripple" characteristic pattern detected in the time-frequency analysis model, a mutation difference in the visceral organ displacement data that conforms to the characteristic of inertial micro-displacement of visceral organs during a fall, and a characteristic tremor signal conforming to the range of 9-13 Hz in the muscle tremor map to comprehensively determine whether the target person is in a fallen state, thereby improving the accuracy and reliability of detection.

[0056] Embodiment 2 Please refer to Figure 5 , the present invention also provides a fall detection device for a smart toilet, including: An information acquisition module; used to detect the height of the target person and the phase angle of the target person in real time through a microwave radar located on the smart toilet; detect whether the seat ring is occupied or not through the seat sensor on the smart toilet; A fall determination module; after the seat sensor detects that the seat ring changes from being occupied to unoccupied, if the height of the target person monitored by the microwave radar is below the preset height, and this state lasts for more than the preset time, and at the same time the target person stays within the preset phase angle range, it is determined that the target person is in a fallen state.

[0057] Furthermore, the fall detection device further includes at least one of the following devices: A first information processing module, by acquiring height values of the target person continuously collected at fixed time intervals to obtain a real-time height measurement value sequence; for the real-time height measurement value sequence, calculate the difference between adjacent height values, perform normalization processing on the difference at fixed time intervals to generate a data sequence of height change rates; extract first derivative data from the data sequence of height change rates, and use numerical differentiation methods to calculate the change amount between adjacent rates to obtain a derivative sequence reflecting the characteristics of the speed of movement; based on the analysis of the derivative sequence, obtain the rate change trend of the target person's movement and determine whether there is a sudden rate anomaly point.

[0058] A second information processing module, by filtering the acquired phase angle of the target person and performing feature extraction to calculate the angular velocity change rate of the phase angle; based on the angular velocity change rate of the phase angle, determine whether there is a sudden angular velocity change anomaly point.

[0059] The third information processing module obtains a sequence of real-time width measurement values by acquiring the horizontal width values of the target person continuously recorded at fixed time intervals; for the sequence of real-time width measurement values, calculates the differences between adjacent horizontal width values, normalizes the differences at fixed time intervals, and generates a data sequence of the horizontal width change rate; based on the horizontal width change rate, determines whether there are sudden abnormal points in the horizontal width change.

[0060] Among them, the fall determination module is also used to determine whether the target person is in a falling state by combining at least one of the sudden rate abnormal points, sudden angular velocity change abnormal points, and sudden horizontal width change abnormal points.

[0061] In addition, the fall detection device may further include modules corresponding to other steps in Embodiment 1 that can assist in determining whether the target person is in a falling state. For example Figure 6 As shown, the fall detection device may further include one or more of a fourth information processing device, a fifth information processing device, and a sixth information processing device. When using the fifth information processing device, the information acquisition device further includes acquiring the displacement data of the internal organs of the target person in real time based on a microwave radar; when using the sixth information processing device, the information acquisition device further includes acquiring the tremor signals generated by the muscle activities of the target person in real time based on a microwave radar.

[0062] Among them, the fourth information processing device is used to extract and process the Doppler micro-motion feature data in the signal of the microwave radar, and construct a time-frequency analysis model of the Doppler frequency shift changing with time, and identify the special "motion ripples" generated at the moment of falling from it; the special "motion ripples" are characterized by a feature pattern with a specific frequency range, duration, and energy distribution that appears in the time-frequency analysis model, and this pattern is different from the time-frequency characteristics when the target person is not at the moment of falling.

[0063] The fifth information processing device is used to pre-establish an organ motion baseline database, which contains the motion feature data of the internal organs of the target person in a normal state; compare and analyze the internal organ displacement data with the data in the organ motion baseline database, and confirm whether there are sudden differences between the internal organ displacement data and the data in the organ motion baseline database and whether the sudden differences conform to the characteristics of the inertial micro-displacement of the internal organs during a fall.

[0064] The sixth information processing device is used to process and analyze the tremor signal to generate a muscle tremor map; compare the generated muscle tremor map with the characteristic tremor signals in the preset range of 9 - 13 Hz to determine whether there are characteristic tremor signals in the muscle tremor map that conform to the range of 9 - 13 Hz.

[0065] When the fourth information processing device is adopted, the determination condition of the fall determination device further includes determining that the target person is in a fallen state by combining the signals detected in the time-frequency analysis model that conform to the special "motion ripple" characteristic pattern. When the fifth information processing device is adopted, the determination condition of the fall determination device further includes determining that the target person is in a fallen state by combining the mutation difference in the visceral organ displacement data that conforms to the characteristic of the inertial micro-displacement of the visceral organs during a fall. When the sixth information processing device is adopted, the determination condition of the fall determination device further includes determining that the target person is in a fallen state by combining the characteristic tremor signals within the range of 9-13 Hz present in the muscle tremor map.

[0066] It should be noted that the specific implementation manners of the above-mentioned various modules can be referred to those described in Embodiment 1, and will not be elaborated herein.

[0067] Embodiment 3 The present invention further provides an intelligent toilet, which adopts the fall detection method for an intelligent toilet described in Embodiment 1 above, or adopts the fall detection device for an intelligent toilet described in Embodiment 2 above. Specifically, it can be referred to the above-mentioned embodiments and will not be elaborated herein.

[0068] Among them, the intelligent toilet includes, but is not limited to, a toilet body, a microwave radar, a seating sensor, a seat ring / cover plate rotation detection device, a control system, and an alarm device, etc. The microwave radar, the seating sensor, the seat ring / cover plate rotation detection device, and the alarm device are respectively electrically connected to the control system. The microwave radar, the seating sensor, and the seat ring / cover plate rotation detection device continuously collect data and transmit it to the control system. The control system analyzes and processes the data according to the above-mentioned information processing method to determine whether there are abnormal points. At the same time, a fall determination is made according to the processing result. If it is determined that the target person is in a fallen state, the alarm device is triggered to notify relevant personnel. At the same time, during the detection process, if the seat ring / cover plate interference condition is met, the sensitivity of the microwave radar is adjusted in a timely manner.

[0069] Through the above settings, the fall detection function can be effectively realized, and the interference of the movable cover plate to the microwave radar can be avoided, providing reliable safety protection for users.

[0070] In addition, those skilled in the art should understand that although there are many problems in the prior art, each embodiment or technical solution of the present invention can be improved in only one or several aspects, and it is not necessary to solve all the technical problems listed in the prior art or the background art at the same time. Those skilled in the art should understand that the content not mentioned in a claim should not be regarded as a limitation to that claim.

[0071] Although terms such as information acquisition step, fall determination step, first information processing step, second information processing step, third information processing step, rate anomaly point, angular velocity change anomaly point, horizontal width change anomaly point, etc. are used more frequently in this article, the possibility of using other terms is not excluded. The use of these terms is only for more conveniently describing and explaining the essence of the present invention; interpreting them as any additional limitation is contrary to the spirit of the present invention; the terms "first", "second", etc. (if any) in the specification, claims and the above-mentioned drawings of the embodiments of the present invention are used to distinguish similar objects and do not necessarily have to be used to describe a specific order or sequence.

[0072] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A fall detection method for an intelligent toilet, characterized in that: The following steps are involved: Information acquisition step: detecting the height and phase angle of the target person in real time by a microwave radar located on the smart toilet; detecting whether there is anyone sitting on the seat ring by a seat sensor on the smart toilet; Fall determination step: when the seating sensor detects that the seat changes from being occupied to being unoccupied, if the height of the target person monitored by the microwave radar is below a preset height, and the duration of this state exceeds a preset time, and the target person stays in a preset phase angle interval, it is determined that the target person is in a falling state.

2. The fall detection method for an intelligent toilet according to claim 1, characterized in that: In the information acquisition step, the millimeter wave radar is placed in the electric control box of the smart toilet, and detects the height and phase angle of the target person at a detection range angle of 120 degrees; the electric control box is located behind the base of the smart toilet and is more than the preset height from the ground.

3. The fall detection method for an intelligent toilet according to claim 1, characterized in that: The method also includes a first information processing step; obtaining a real-time height measurement value sequence by recording and continuously collecting the height values ​​of the target person at fixed time intervals; calculating the difference between adjacent height values ​​for the real-time height measurement value sequence, normalizing the difference at fixed time intervals, and generating a data sequence of height change rate; extracting first-order derivative data from the data sequence of height change rate, calculating the change between adjacent rates using a numerical differentiation method, and obtaining a derivative sequence reflecting the speed characteristics of the action; obtaining the rate change trend of the target person's action based on the derivative sequence analysis, and judging whether there is a sudden rate abnormality point; In the fall determination step, if there is a sudden rate abnormal point, after the seat sensor detects that the seat circle changes from someone sitting to no one sitting, the microwave radar monitors that the height of the target person is below the preset height and the duration reaches more than the preset time, and the target person stays in the interval of the preset phase angle, it is determined that the target person is in a fall state; And / or, it also includes a second information processing step, filtering the acquired target person's phase angle and performing feature extraction to calculate the angular velocity change rate of the phase angle; based on the angular velocity change rate of the phase angle, determining whether there is a sudden angular velocity change abnormal point; In the fall determination step, if there is a sudden abnormal point of angular velocity change, the seat sensor detects that the seat changes from someone sitting to no one sitting, and after the sudden abnormal point of angular velocity change, the microwave radar detects that the height of the target person is below the preset height and lasts for more than the preset time, and the target person stays in the interval of the preset phase angle, it is determined that the target person is in a fall state; And / or, in the information acquisition step, it also includes acquiring the horizontal width of the target person in real time based on the microwave radar; and also includes a third information processing step, by acquiring the horizontal width values ​​of the target person recorded and continuously collected at fixed time intervals, to obtain a real-time width measurement value sequence; for the real-time width measurement value sequence, the difference between adjacent horizontal width values ​​is calculated, and the difference is normalized at fixed time intervals to generate a data sequence of horizontal width change rate; based on the horizontal width change rate, it is determined whether there is a sudden abnormal point of horizontal width change; In the fall determination step, if there is a sudden abnormal point in the horizontal width change, the seat sensor detects that the seat changes from someone sitting on it to no one sitting on it, and after the sudden abnormal point in the horizontal width change, the microwave radar monitors that the height of the target person is below the preset height and the duration reaches more than the preset time, and the target person stays in the preset phase angle interval, it is determined that the target person is in a fall state.

4. The fall detection method for an intelligent toilet according to claim 1, characterized in that: The vertical area where the smart toilet is located is set as an interference area. When a microwave radar signal is detected in the interference area and the seat ring and / or the cover plate rotates, the sensitivity of the microwave radar is reduced.

5. The fall detection method for an intelligent toilet according to claim 1, characterized in that: In the fall determination step, when the seating sensor detects that the seat changes from being occupied to being unoccupied, if the height of the target person does not reach the highest height range of the target person detected during the rising process, the height of the target person monitored by the microwave radar is below the preset height, and the duration of this state exceeds the preset time, and the target person stays in the preset phase angle range, it is determined that the target person is in a fall state.

6. The fall detection method for an intelligent toilet according to any one of claims 1 to 5, characterized in that: The method further includes a fourth information processing step, which is used to extract and process Doppler micro-motion feature data in the signal of the microwave radar, and construct a time-frequency analysis model of Doppler frequency shift changing with time, thereby identifying the special "motion ripples" generated at the moment of falling; the special "motion ripples" are manifested as characteristic patterns with a specific frequency range, duration and energy distribution that appear in the time-frequency analysis model, and the pattern is different from the time-frequency characteristics of the target person when he is not at the moment of falling; In the fall determination step, when the seating sensor detects that the seat changes from someone sitting on it to no one sitting on it, if the height of the target person monitored by the microwave radar is below the preset height, and the duration of this state reaches more than the preset time, and the target person stays in the interval of the preset phase angle, and a signal that conforms to the special "motion ripple" characteristic pattern is detected in the time-frequency analysis model, it is determined that the target person is in a fall state.

7. The fall detection method for an intelligent toilet according to any one of claims 1 to 5, characterized in that: The information acquisition step also includes real-time collection of visceral organ displacement data of the target person based on microwave radar; and also includes a fifth information processing step, pre-establishing an organ motion baseline database, which contains motion characteristic data of the visceral organs of the target person in a normal state; comparing and analyzing the visceral organ displacement data with the data in the organ motion baseline database, and confirming whether there is a mutation difference between the visceral organ displacement data and the data in the organ motion baseline database, and whether the mutation difference conforms to the characteristics of the inertial micro-displacement of the visceral organs when falling; In the fall determination step, when the seat sensor detects that the seat ring changes from being occupied to being unoccupied, if the height of the target person monitored by the microwave radar is below a preset height, and the state lasts for more than a preset time, and the target person stays in a preset phase angle interval, and the internal organ displacement data has a sudden change difference that meets the characteristics of the inertial micro-displacement of the internal organs during a fall, it is determined that the target person is in a fall state; And / or, in the information acquisition step, it also includes collecting tremor signals generated by muscle activities of the target person in real time based on microwave radar; and also includes a sixth information processing step for processing and analyzing the tremor signals to generate a muscle tremor spectrum; comparing the generated muscle tremor spectrum with a preset characteristic tremor signal in the range of 9 to 13 Hz to determine whether there is a characteristic tremor signal in the range of 9 to 13 Hz in the muscle tremor spectrum; In the fall determination step, when the seating sensor detects that the seat changes from being occupied to being unoccupied, if the height of the target person monitored by the microwave radar is below a preset height and the duration of this state exceeds a preset time, and the target person stays in a preset phase angle interval, and there is a characteristic tremor signal in the range of 9 to 13 Hz in the muscle tremor spectrum, it is determined that the target person is in a fall state.

8. A fall detection device for an intelligent toilet, characterized in that: include: Information acquisition module; It is used to detect the height and phase angle of the target person in real time through the microwave radar located on the smart toilet; and to detect whether there is anyone sitting on the seat ring through the seat sensor on the smart toilet; Fall determination module: when the seating sensor detects that the seat changes from being occupied to being unoccupied, if the height of the target person monitored by the microwave radar is below a preset height, and the duration of this state exceeds a preset time, and the target person stays in a preset phase angle interval, it is determined that the target person is in a fall state.

9. The fall detection device for an intelligent toilet according to claim 8, characterized in that: Also includes: The first information processing module obtains the height values ​​of the target person by recording and continuously collecting the height values ​​at fixed time intervals to obtain a real-time height measurement value sequence; for the real-time height measurement value sequence, the difference between adjacent height values ​​is calculated, and the difference is normalized at fixed time intervals to generate a data sequence of height change rate; first-order derivative data is extracted from the data sequence of height change rate, and the change amount between adjacent rates is calculated by numerical differentiation method to obtain a derivative sequence reflecting the speed characteristics of the action; based on the derivative sequence analysis, the rate change trend of the target person's action is obtained to determine whether there is a sudden rate abnormal point; The second information processing module performs filtering processing on the acquired target person's phase angle and performs feature extraction to calculate the angular velocity change rate of the phase angle; based on the angular velocity change rate of the phase angle, determines whether there is a sudden abnormal point of angular velocity change; The third information processing module obtains the horizontal width values ​​of the target person by recording and continuously collecting them at fixed time intervals to obtain a real-time width measurement value sequence; for the real-time width measurement value sequence, the difference between adjacent horizontal width values ​​is calculated, and the difference is normalized at fixed time intervals to generate a data sequence of horizontal width change rate; based on the horizontal width change rate, it is determined whether there is a sudden abnormal point of horizontal width change; The fall determination module is further used to determine whether the target person is in a fall state by combining at least one of a sudden rate abnormality point, a sudden angular velocity change abnormality point, and a sudden horizontal width change abnormality point.

10. An intelligent toilet, characterized in that: A fall detection method for an intelligent toilet as described in any one of claims 1 to 7 is adopted, or a fall detection device for an intelligent toilet as described in claim 8 or 9 is adopted.

Images