A smart bed driving system based on human posture recognition

Abstract

The present invention relates to the field of intelligent drive technology, and in particular to an intelligent bed drive system based on human posture recognition, the system comprising: a data acquisition module, a data processing module, a drive control module, an execution operation module and a feedback correction module. The present invention classifies and marks the real-time action amplitude by calculating the actual turning frequency, and determines the preliminary staging result according to the actual turning frequency and the classification marking result; determines the corresponding staging evaluation base value based on the preliminary staging result, screens the target pressure parameter according to the staging evaluation base value combined with the target position posture, obtains the key characteristic parameter and generates the corresponding key characteristic change curve; identifies the key characteristic change curve to determine the actual fluctuation situation, determines the actual target mode according to the actual fluctuation situation, and sends a corresponding control signal based on the actual target mode; performs corresponding action adjustment according to the control signal, and sends a corresponding completion signal when the action adjustment is completed.

Description

Technical Field

[0001] The present invention relates to the field of intelligent driving technology, and in particular to an intelligent bed driving system based on human body posture recognition. Background Art

[0002] Existing smart beds do not use advanced posture recognition technology, which makes it impossible to accurately capture the user's subtle posture changes; they are not enough to monitor the user's posture comprehensively and in real time; the posture data monitored in real time needs to undergo complex processing and analysis, and cannot accurately judge the user's sleeping state and adjustment needs, resulting in an inability to respond in a timely and accurate manner; therefore, existing smart beds still have some technical limitations and practical application problems in real-time monitoring of user posture and adjusting the drive.

[0003] Chinese patent application publication number CN107136815A discloses a remote control system based on a smart bed, including a mattress, whose upper surface is provided with a plurality of pressure sensors arranged in a grid; a server, which is connected to the pressure sensors; and a user terminal, which is used to input medical order data and transmit the medical order data to the server; wherein the server determines whether the user needs assistance in changing the lying position based on the medical order data and the pressure data of the pressure sensor, and if the user needs assistance in changing the lying position, the server starts the driving part of the mattress to assist the user in changing the lying position.

[0004] This shows that current smart beds are unable to achieve accurate and timely adjustment and drive for users in different sleeping states. Summary of the invention

[0005] To this end, the purpose of the present invention is to provide a smart bed driving system based on human posture recognition, which is used to overcome the problem that current smart beds cannot achieve accurate and timely adjustment and driving of users in different sleeping states.

[0006] To achieve the above object, the present invention provides an intelligent bed driving system based on human posture recognition, comprising:

[0007] A data acquisition module is used to monitor the target in real time to obtain real-time posture data, and collect the target posture data in real time at a preset initial time interval;

[0008] A data processing module, which is used to determine the preliminary posture staging result of the target to be monitored according to the real-time monitoring result, analyze the target posture data based on the preliminary staging result, determine the key characteristic parameters of the target to be monitored according to the primary analysis result, and generate a corresponding key characteristic change curve according to the key characteristic parameters;

[0009] A drive control module, which is used to identify the key feature change curve to determine the actual fluctuation situation, determine the actual target mode according to the actual fluctuation situation, and issue a corresponding control signal based on the actual target mode;

[0010] An execution operation module, which is used to receive the control signal, perform corresponding action adjustment according to the control signal, and send a corresponding completion signal when the action adjustment is completed;

[0011] A feedback correction module, which collects the actual posture data of the target to be monitored in real time at intermediate time intervals based on the completion signal, performs secondary analysis based on the actual posture data combined with the real-time monitoring result, determines whether the action adjustment is qualified based on the secondary analysis result, and starts the feedback correction mode when the action adjustment is unqualified;

[0012] The intermediate time interval is equal to 1 / 3 of the initial time interval.

[0013] Further, the data acquisition module includes: a video monitoring unit and a timing unit;

[0014] The video monitoring unit is used to monitor the target to be monitored in real time to obtain the real-time turning action and real-time action amplitude of the target to be monitored;

[0015] The timing unit starts timing when the actual duration of the monitored target staying in the monitoring target area meets the timing start condition to obtain the actual monitoring duration;

[0016] Wherein, when the actual stay time is greater than or equal to the preset standard stay time, the timing unit determines that the timing start condition is met; when the actual stay time is less than the standard stay time, the timing unit determines that the timing start condition is not met;

[0017] The real-time posture data includes: the real-time turning over action and the real-time action amplitude.

[0018] Furthermore, the data acquisition module further includes: an image acquisition unit and a plurality of pressure sensors;

[0019] The image acquisition unit acquires a real-time image every time the actual monitoring time reaches the initial time interval to obtain a target position posture;

[0020] The pressure sensors are evenly distributed in the monitoring target area, and collect the target pressure parameters of the target to be monitored in real time at the initial time interval;

[0021] Wherein, the target posture data includes: the target position posture and the target pressure parameter.

[0022] Further, the data processing module includes: a determination unit, a posture analysis unit and a generation unit;

[0023] The determination unit is used to calculate the actual turning frequency according to the real-time turning action and the actual monitoring duration, classify and mark the real-time action amplitude, and determine the preliminary staging result according to the actual turning frequency and the classification and marking result;

[0024] The posture analysis unit determines a corresponding stage evaluation base value based on the preliminary stage evaluation result, and screens the target pressure parameter according to the stage evaluation base value combined with the target position posture to obtain the key characteristic parameter;

[0025] The generating unit is used to generate the corresponding key feature variation curve according to the key feature parameter.

[0026] Further, the determination unit includes: an accumulation subunit and a calculation subunit;

[0027] The accumulating subunit is used to accumulate the real-time turning over actions to obtain the actual turning over times, and obtain the actual monitoring time when the actual turning over times meets the calculation start condition;

[0028] The calculation subunit is used to calculate the actual turning over frequency according to the actual number of turning over times and the actual monitoring time;

[0029] Among them, when the actual number of turning over is greater than or equal to the preset standard number of turning over, the accumulating subunit determines that the calculation start condition is met; when the actual number of turning over is less than the standard number of turning over, the accumulating subunit determines that the calculation start condition is not met.

[0030] Further, the determination unit further comprises: a classification subunit and a marking subunit;

[0031] The classification subunit is used to compare the real-time motion amplitude with a preset standard motion amplitude to determine the record type of the real-time motion amplitude;

[0032] The marking subunit is used to mark corresponding actions according to the classification result of the classification subunit;

[0033] The classification subunit determines the record type according to the real-time amplitude difference and a preset first evaluation value; the real-time amplitude difference is the absolute value of the difference between the real-time action amplitude and the standard action amplitude.

[0034] Further, based on the fact that the real-time amplitude difference is less than or equal to the first evaluation value, the classification subunit determines that the record type is a low-intensity action;

[0035] When the real-time amplitude difference is greater than the first evaluation value and the real-time action amplitude is greater than the standard action amplitude, the classification subunit determines that the record type is a high-intensity action;

[0036] When the real-time amplitude difference is greater than the first evaluation value and the real-time motion amplitude is smaller than the standard motion amplitude, the classification subunit determines that the record type is normal micro-motion.

[0037] Furthermore, the determination unit further comprises: a determination subunit;

[0038] The determination subunit integrates the classification results of the classification subunit based on the actual turning frequency meeting the calculation start condition, determines the actual turning frequency and the number of items in the integration result that meet the single determination condition, and determines the preliminary staging result according to the number of items;

[0039] Among them, when the actual turning over frequency is greater than or equal to the preset standard turning over frequency, the determination subunit determines that the actual turning over frequency meets the single determination condition; when the actual turning over frequency is less than the standard turning over frequency, the determination subunit determines that the actual turning over frequency does not meet the single determination condition.

[0040] Further, the determination subunit calculates an actual strength score according to the integration result, and determines whether the integration situation meets the single determination condition according to the actual strength score and a preset standard strength score;

[0041] Among them, based on the actual strength score being greater than or equal to the standard strength score, the determination subunit determines that the integration situation meets the single determination condition; based on the actual strength score being less than the standard strength score, the determination subunit determines that the integration situation does not meet the single determination condition.

[0042] Further, the posture analysis unit includes: an optimization subunit and a screening subunit;

[0043] The optimization subunit determines an optimization compensation value according to the target position and posture, and optimizes the staged evaluation base value according to the optimization compensation value to obtain a staged evaluation threshold value;

[0044] The screening subunit is used to determine whether the target pressure parameter is abnormal according to the stage evaluation threshold, and determine the target pressure parameter as the key characteristic parameter if it is abnormal;

[0045] Among them, based on the target pressure parameter being greater than or equal to the stage evaluation threshold, the screening subunit determines it as the key feature parameter; based on the target pressure parameter being less than the stage evaluation threshold, the screening subunit determines it as not the key feature parameter.

[0046] Compared with the prior art, the beneficial effects of the present invention are that the actual turning over frequency, classification and marking of movement amplitudes, and integration of this information are used to determine the preliminary staging results of the user; based on the preliminary staging results, the target posture data is deeply analyzed, key feature parameters are extracted, and a key feature change curve is generated; it ensures that the different sleep states of the user can be accurately identified; the actual fluctuation of the user is identified according to the key feature change curve, and the actual target mode is determined; a corresponding control signal is issued, and the state of the smart bed is adjusted by executing the operation module, such as adjusting the hardness and tilt angle of the mattress to adapt to the user's current sleep needs; after the movement adjustment is completed, the control module is used to adjust the state of the smart bed ... calculating the hardness and tilt angle of the mattress to adapt to the user's current sleep needs; after the movement adjustment is completed, the control module is used to adjust the hardness and tilt angle of the mattress to adapt to the user's current sleep needs; after the movement adjustment is completed The user's actual posture data is collected in real time through the feedback correction module and a secondary analysis is performed. If the action adjustment is unsatisfactory, the feedback correction mode will be turned on to further optimize the adjustment strategy. The closed-loop feedback mechanism ensures that the state of the smart bed can be adjusted continuously and accurately. The influence of the user's specific posture (such as side sleeping and supine lying) on ​​the pressure distribution is also considered, and the stage evaluation base value is optimized by optimizing the compensation value to obtain a stage evaluation threshold that is more in line with the user's actual state. The personalized optimization strategy further improves the accuracy of system adjustment and the user's comfort. It realizes precise special case adjustment control of users in different sleeping states, thereby effectively ensuring the user's sleep quality.

[0047] Through the video monitoring unit and pressure sensor, the real-time turning over movement and movement amplitude of the monitored target, as well as the target position posture and pressure parameters can be accurately obtained, providing detailed data support for subsequent analysis; combined with video monitoring and pressure sensor data, the activity mode and state of the monitored target can be more comprehensively understood; the timing unit decides whether to start timing according to the actual stay time of the monitored target in the monitoring target area, which helps to more accurately judge whether the target has entered the sleep preparation state; the setting of the standard stay time can be adaptively adjusted according to the age of the monitored target and actual environmental factors, which improves the flexibility and applicability of the system; defining the monitoring target area as the surface of the smart bed helps to monitor sleep-related activities in a targeted manner, improving the targetedness and efficiency of monitoring; by combining real-time images and pressure data, the state of the monitored target can be analyzed from multiple dimensions, improving the accuracy and reliability of the analysis; data collection at a preset initial time interval ensures that the target posture data can be obtained and fed back in real time; it helps to monitor and analyze sleep quality and provide data support for improving sleep; by monitoring the turning over frequency and movement amplitude, abnormal movements can be discovered in time, realizing accurate monitoring of sleep.

[0048] By calculating the actual turning over frequency, the activity level of the target to be monitored can be evaluated more accurately; classifying and marking the real-time movement amplitude can help identify different movement patterns and intensities, thereby improving the precision of data analysis; based on the classification results of turning over frequency and movement amplitude, the state of the target can be preliminarily judged to provide a basis for subsequent processing; by setting evaluation base values ​​for different stages, more targeted analysis can be performed according to the specific state of the target; combined with the target position posture and pressure parameters, the posture analysis unit can screen out key feature parameters, which helps to extract the most valuable information for further evaluation and decision-making; by generating key feature change curves, it helps to reveal the activity trends and patterns of the target to be monitored, realizes the automated process from data collection to analysis, and improves processing efficiency; through precise calculation and screening processes, it can provide more reliable data analysis results for evaluating the state of the target to be monitored; it can process and analyze data in real time and provide timely feedback on the state of the target to be monitored.

[0049] By accumulating turning movements in real time and calculating the actual number of turning movements, accurate capture of turning movements is ensured; when the actual number of turning movements reaches the preset standard, the actual turning frequency is further calculated, which provides important data support for evaluating the sleep quality of the monitored target; the adaptive adjustment of the standard enhances the flexibility and practicality of the system, enabling it to more accurately adapt to the needs of different monitored objects; by comparing the real-time movement amplitude with the standard movement amplitude, the movements are subdivided into low-intensity movements, high-intensity movements and normal micro-movements. This detailed classification helps to more comprehensively understand the sleep status of the monitored target; marking the movements according to the classification results is convenient for subsequent data analysis and processing, and improves the utilization and accuracy of the monitoring data; through the automated data collection and processing process, manual intervention is reduced and the monitoring efficiency is improved; at the same time, based on the precise number of turning and movement classification, more accurate monitoring results can be provided, which helps to detect abnormal situations in time and take corresponding measures; for individuals who need continuous monitoring, timely and accurate monitoring data can be provided to help better understand the physical condition, thereby enhancing user experience and safety; through continuous monitoring of turning frequency and movement amplitude, it can help identify potential health problems or risk factors, such as sleep disorders, reduced activity, etc., so that timely intervention measures can be taken.

[0050] By combining multiple dimensions such as the actual number of turning over, the actual turning over frequency and the actual intensity score for judgment, it can more accurately reflect the sleep status of the monitored target; the multi-dimensional comprehensive evaluation method can capture subtle changes better than a single indicator, thereby improving the accuracy and reliability of the evaluation; it can automatically integrate the classification results of the classification sub-units and make judgments based on preset conditions, thereby realizing the automation and intelligence of the evaluation process; it reduces the burden of manual evaluation, reduces human errors and improves evaluation efficiency; by making staged judgments based on the individual's actual turning over frequency and intensity score, it can provide each person with personalized evaluation results, which helps to more accurately understand the sleep status of the monitored target and provide strong support for subsequent interventions.

[0051] By introducing the optimized compensation value, the stage evaluation base value can be adjusted according to different target positions and postures, so as to obtain a stage evaluation threshold that is more in line with the actual state of the target; the personalized adjustment significantly improves the evaluation accuracy of the target posture and pressure parameters; it can adapt to the differences between different individuals, such as weight, body shape, and muscle tension, and reflect the impact of these differences on posture and pressure distribution by adjusting the optimized compensation value; it enhances the adaptability and practicality in different users and application scenarios; the target pressure parameters are monitored in real time according to the stage evaluation threshold, and once an abnormality is found, it is identified as a key characteristic parameter; it helps to timely discover potential health problems or risks and provide support for subsequent interventions; the collaborative work of the optimization subunit and the screening subunit can more accurately evaluate the sleep state of the monitored target. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a schematic diagram of the structure of an intelligent bed driving system based on human posture recognition according to an embodiment of the present invention;

[0053] Figure 2 This is a structural schematic diagram of a data acquisition module in a smart bed driving system based on human posture recognition according to an embodiment of the present invention;

[0054] Figure 3 This is a structural schematic diagram of a data processing module in a smart bed driving system based on human posture recognition according to an embodiment of the present invention;

[0055] Figure 4 It is a structural schematic diagram of a determination unit in a smart bed driving system based on human posture recognition according to an embodiment of the present invention;

[0056] The figure includes: 1-data acquisition module, 2-data processing module, 3-drive control module, 4-execution operation module and 5-feedback correction module. DETAILED DESCRIPTION

[0057] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0058] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0059] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is merely for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0060] See also Figure 1-Figure 4 As shown, Figure 1 This is a schematic diagram of the structure of an intelligent bed driving system based on human posture recognition according to an embodiment of the present invention; Figure 2 This is a structural schematic diagram of a data acquisition module in a smart bed driving system based on human posture recognition according to an embodiment of the present invention; Figure 3 This is a structural schematic diagram of a data processing module in a smart bed driving system based on human posture recognition according to an embodiment of the present invention; Figure 4 It is a structural schematic diagram of a determination unit in a smart bed driving system based on human posture recognition according to an embodiment of the present invention.

[0061] The present invention provides an intelligent bed driving system based on human posture recognition, comprising:

[0062] A data acquisition module 1 is used to monitor the target to be monitored in real time to obtain real-time posture data, and collect the target posture data in real time at a preset initial time interval;

[0063] A data processing module 2 is used to determine the preliminary posture staging result of the target to be monitored according to the real-time monitoring result, analyze the target posture data based on the preliminary staging result, determine the key characteristic parameters of the target to be monitored according to the analysis result, and generate a corresponding key characteristic change curve according to the key characteristic parameters;

[0064] A drive control module 3, which is used to identify the key characteristic change curve to determine the actual fluctuation situation, determine the actual target mode according to the actual fluctuation situation, and issue a corresponding control signal based on the actual target mode;

[0065] An execution operation module 4 is used to receive the control signal, perform corresponding action adjustment according to the control signal, and send a corresponding completion signal when the action adjustment is completed;

[0066] A feedback correction module 5, which collects the actual posture data of the target to be monitored in real time at intermediate time intervals based on the completion signal, performs secondary analysis based on the actual posture data combined with the real-time monitoring results, determines whether the action adjustment is qualified based on the secondary analysis results, and starts the feedback correction mode when the action adjustment is unqualified;

[0067] The intermediate time interval is equal to 1 / 3 of the initial time interval.

[0068] In this embodiment, the initial time interval is set to 10 minutes to 15 minutes, and is adaptively selected and adjusted according to the age of the target to be monitored and actual environmental factors.

[0069] Specifically, the embodiment of the present invention calculates the actual turning frequency, classifies and marks the action amplitude, and integrates this information to determine the preliminary staging result of the user. Based on the preliminary staging result, the target posture data is deeply analyzed, the key feature parameters are extracted, and the key feature change curve is generated; it ensures that the different sleep states of the user can be accurately identified; the actual fluctuation of the user is identified according to the key feature change curve, and the actual target mode is determined; a corresponding control signal is issued, and the state of the smart bed is adjusted by executing the operation module, such as adjusting the hardness and tilt angle of the mattress, so as to adapt to the current sleep needs of the user; after the action adjustment is completed, the actual posture data of the user is collected in real time through the feedback correction module, and a secondary analysis is performed; if the action adjustment is unqualified, the feedback correction mode will be turned on to further optimize the adjustment strategy; the closed-loop feedback mechanism ensures that the state of the smart bed can be continuously and accurately adjusted; the influence of the user's specific posture (such as side lying and supine lying) on ​​the pressure distribution is also considered, and the stage evaluation base value is optimized by optimizing the compensation value to obtain a stage evaluation threshold that is more in line with the actual state of the user; the personalized optimization strategy further improves the accuracy of system adjustment and the comfort of the user; and the precise special case adjustment control of the user in different sleep states is realized, thereby effectively ensuring the sleep quality of the user.

[0070] Through real-time monitoring and data collection at preset initial time intervals, changes in the target's posture can be quickly captured, providing real-time information for subsequent processing; continuous data collection helps to observe and analyze the target's long-term behavior patterns; through preliminary staging and one-time analysis, the key characteristic parameters of the target can be more accurately identified, thus providing a basis for deeper analysis; the change curve generated based on the key characteristic parameters helps to specifically identify the target's behavioral characteristics; the actual fluctuation situation can be identified based on the key characteristic change curve, the target mode can be intelligently determined, and the corresponding control signal can be issued; the issuance of the control signal is based on the actual target mode, which makes it have better environmental adaptability and operational flexibility; the action is automatically adjusted according to the control signal , which improves the degree of automation of operations; the completion signal issued after the action adjustment is completed helps to understand the operation status in time and improve the overall operation efficiency; it can perform secondary analysis based on the actual posture data and real-time monitoring results to ensure the accuracy of the action adjustment; when the action adjustment is unqualified, the feedback correction mode is turned on, which helps self-optimization and continuously improves the adjustment effect; the intermediate time interval is set to 1 / 3 of the initial time interval, which means that during the action adjustment, the frequency of data collection is higher, which helps to monitor and adjust the target posture more finely; improve the real-time and continuity of monitoring, enhance the accuracy and pertinence of analysis, realize intelligent response and automated operation, and improve the overall performance and efficiency through closed-loop control and self-optimization.

[0071] Specifically, the data acquisition module 1 in this embodiment includes: a video monitoring unit and a timing unit;

[0072] The video monitoring unit is used to monitor the target to be monitored in real time to obtain the real-time turning action and real-time action amplitude of the target to be monitored;

[0073] The timing unit starts timing when the actual duration of the monitored target staying in the monitoring target area meets the timing start condition to obtain the actual monitoring duration;

[0074] Wherein, when the actual stay time is greater than or equal to the preset standard stay time, the timing unit determines that the timing start condition is met; when the actual stay time is less than the standard stay time, the timing unit determines that the timing start condition is not met;

[0075] The monitoring target area is a space or surface area that can accurately monitor and record the activities of the target to be monitored. In this embodiment, the monitoring target area is the surface of the smart bed;

[0076] In the specific implementation process, the standard stay time is set to 5min-8min, and the standard stay time is used to measure whether the target to be monitored is ready for sleep; and adaptive selection and adjustment are made according to the age of the target to be monitored and the actual environmental factors; and adaptive selection and adjustment are made according to the age of the target to be monitored and the actual environmental factors;

[0077] The real-time posture data includes: the real-time turning over action and the real-time action amplitude;

[0078] The data acquisition module 1 also includes: an image acquisition unit and a plurality of pressure sensors;

[0079] The image acquisition unit acquires a real-time image every time the actual monitoring time reaches the initial time interval to obtain a target position posture;

[0080] The pressure sensors are evenly distributed in the monitoring target area, and collect the target pressure parameters of the target to be monitored in real time at the initial time interval;

[0081] Wherein, the target posture data includes: the target position posture and the target pressure parameter.

[0082] Specifically, the embodiment of the present invention can accurately obtain the real-time turning over movement and movement amplitude of the target to be monitored, as well as the target position posture and pressure parameters through the video monitoring unit and the pressure sensor, providing detailed data support for subsequent analysis; combining the video monitoring and pressure sensor data, the activity mode and state of the target to be monitored can be more comprehensively understood; the timing unit decides whether to start the timing according to the actual stay time of the target to be monitored in the monitoring target area, which helps to more accurately judge whether the target has entered the sleep preparation state; the setting of the standard stay time can be adaptively adjusted according to the age of the target to be monitored and the actual environmental factors, thereby improving the flexibility and applicability of the system; defining the monitoring target area as the surface of the smart bed helps to monitor sleep-related activities in a targeted manner, thereby improving the targetedness and efficiency of monitoring; by combining real-time images and pressure data, the state of the target to be monitored can be analyzed from multiple dimensions, thereby improving the accuracy and reliability of the analysis; data collection is performed at a preset initial time interval, thereby ensuring that the target posture data can be obtained and fed back in real time; it helps to monitor and analyze sleep quality, thereby providing data support for improving sleep; by monitoring the turning over frequency and movement amplitude, abnormal movements can be discovered in a timely manner, thereby achieving accurate monitoring of sleep.

[0083] Specifically, the data processing module 2 in this embodiment includes: a determination unit, a posture analysis unit and a generation unit;

[0084] The determination unit is used to calculate the actual turning frequency according to the real-time turning action and the actual monitoring duration, classify and mark the real-time action amplitude, and determine the preliminary staging result according to the actual turning frequency and the classification and marking result;

[0085] The posture analysis unit determines a corresponding stage evaluation base value based on the preliminary stage evaluation result, and screens the target pressure parameter according to the stage evaluation base value combined with the target position posture to obtain the key characteristic parameter;

[0086] The generating unit is used to generate the corresponding key feature variation curve according to the key feature parameter.

[0087] Specifically, the embodiments of the present invention can more accurately evaluate the activity level of the target to be monitored by calculating the actual turning over frequency; classifying and marking the real-time movement amplitude is helpful to identify different movement patterns and intensities, thereby improving the precision of data analysis; based on the classification results of turning over frequency and movement amplitude, it is possible to preliminarily judge the state of the target and provide a basis for subsequent processing; by setting evaluation base values ​​for different stages, it is possible to perform more targeted analysis according to the specific state of the target; in combination with the target position posture and pressure parameters, the posture analysis unit can screen out key feature parameters, which is helpful to extract the most valuable information for further evaluation and decision-making; by generating key feature change curves, it is helpful to reveal the activity trends and patterns of the target to be monitored, realize the automated process from data collection to analysis, and improve processing efficiency; through precise calculation and screening processes, it can provide more reliable data analysis results for evaluating the state of the target to be monitored; it can process and analyze data in real time and provide timely feedback on the state of the target to be monitored.

[0088] Specifically, the determination unit in this embodiment includes: an accumulation subunit and a calculation subunit;

[0089] The accumulating subunit is used to accumulate the real-time turning over actions to obtain the actual turning over times, and obtain the actual monitoring time when the actual turning over times meets the calculation start condition;

[0090] The calculation subunit is used to calculate the actual turning over frequency according to the actual number of turning over times and the actual monitoring time;

[0091] Wherein, when the actual number of turning over is greater than or equal to the preset standard number of turning over, the accumulating subunit determines that the calculation start condition is met; when the actual number of turning over is less than the standard number of turning over, the accumulating subunit determines that the calculation start condition is not met;

[0092] In this embodiment, the standard number of turning over is set to 5-10 times; and is adaptively selected and adjusted according to the age of the target to be monitored and the actual environmental factors;

[0093] The determination unit further comprises: a classification subunit and a marking subunit;

[0094] The classification subunit is used to compare the real-time motion amplitude with a preset standard motion amplitude to determine the record type of the real-time motion amplitude;

[0095] The marking subunit is used to mark corresponding actions according to the classification result of the classification subunit;

[0096] Wherein, the classification subunit determines the record type according to the real-time amplitude difference and the preset first evaluation value; the real-time amplitude difference is the absolute value of the difference between the real-time action amplitude and the standard action amplitude;

[0097] The real-time amplitude difference=|real-time action amplitude-standard action amplitude|;

[0098] When the real-time amplitude difference is less than or equal to the first evaluation value, the classification subunit determines that the record type is a low-intensity action;

[0099] When the real-time amplitude difference is greater than the first evaluation value and the real-time action amplitude is greater than the standard action amplitude, the classification subunit determines that the record type is a high-intensity action;

[0100] When the real-time amplitude difference is greater than the first evaluation value and the real-time motion amplitude is less than the standard motion amplitude, the classification subunit determines that the record type is a normal micro-motion;

[0101] In this embodiment, the standard movement amplitude is used to measure the movement range of the target to be monitored; the first evaluation value is the determination threshold of the record type of the target to be monitored; the standard movement amplitude is set to 70cm 3 -90cm 3 ; The first evaluation value is set to 10cm 3 -15cm 3 ; and make adaptive selection and adjustments based on the age of the target to be monitored and actual environmental factors.

[0102] Specifically, the embodiment of the present invention ensures accurate capture of turning over movements by accumulating turning over movements in real time and calculating the actual number of turning over times; when the actual number of turning over times reaches the preset standard, the actual turning over frequency is further calculated, which provides important data support for evaluating the sleep quality of the target to be monitored; the adaptive adjustment of the standard enhances the flexibility and practicality of the system, enabling it to more accurately adapt to the needs of different monitored objects; by comparing the real-time movement amplitude with the standard movement amplitude, the movements are subdivided into low-intensity movements, high-intensity movements and normal micro-movements. This detailed classification helps to more comprehensively understand the sleep status of the target to be monitored; the movements are marked according to the classification results, which is convenient for Subsequent data analysis and processing improve the utilization and accuracy of monitoring data; through automated data collection and processing processes, manual intervention is reduced and monitoring efficiency is improved; at the same time, based on precise turning times and action classification, it can provide more accurate monitoring results, which helps to detect abnormal situations in time and take corresponding measures; for individuals who need continuous monitoring, it can provide timely and accurate monitoring data to help better understand the physical condition, thereby enhancing user experience and safety; through continuous monitoring of turning frequency and movement amplitude, it can help identify potential health problems or risk factors, such as sleep disorders, reduced activity, etc., so that timely intervention measures can be taken.

[0103] Specifically, the determination unit in this embodiment further includes: a determination subunit;

[0104] The determination subunit integrates the classification results of the classification subunit based on the actual turning frequency meeting the calculation start condition, determines the actual turning frequency and the number of items in the integration result that meet the single determination condition, and determines the preliminary staging result according to the number of items;

[0105] Wherein, when the actual turning over frequency is greater than or equal to the preset standard turning over frequency, the determination subunit determines that the actual turning over frequency meets the single determination condition; when the actual turning over frequency is less than the standard turning over frequency, the determination subunit determines that the actual turning over frequency does not meet the single determination condition;

[0106] The determination subunit calculates an actual strength score according to the integration result, and determines whether the integration situation meets the single determination condition according to the actual strength score and a preset standard strength score;

[0107] Among them, based on the actual strength score being greater than or equal to the standard strength score, the determination subunit determines that the integration situation meets the single determination condition; based on the actual strength score being less than the standard strength score, the determination subunit determines that the integration situation does not meet the single determination condition.

[0108] Specifically, the determination subunit in this embodiment determines the preliminary staging result according to the number of items;

[0109] If the number of items is 0, the determination subunit determines that the preliminary staging result is a normal stage;

[0110] If the number of items is 1, the determination subunit determines that the preliminary staging result is a mild abnormal stage;

[0111] If the number of items is 2, the determination subunit determines that the preliminary staging result is a severe abnormal stage;

[0112] The determination subunit integrates the classification results of the classification subunit;

[0113] Integrate the real-time movement amplitudes of the low-intensity movements, to obtain a first set;

[0114] Integrate the real-time motion amplitudes of the high-intensity motions, to obtain a second set;

[0115] Read a first quantity in the first set, and read a second quantity in the second set;

[0116] The determination subunit calculates the actual strength score according to the first number and the second number;

[0117] The actual strength score=the first quantity×a1+the second quantity×a2;

[0118] Wherein, a1 is a first calculation compensation parameter of the first quantity for the actual strength score; a2 is a second calculation compensation parameter of the second quantity for the actual strength score;

[0119] In this embodiment, the first calculation compensation parameter is set to 0.3; the second calculation compensation parameter is set to 0.8; the standard intensity score is set to 4-6; and adaptive selection and adjustment are performed according to the age of the target to be monitored and actual environmental factors.

[0120] Specifically, the embodiments of the present invention can more accurately reflect the sleep status of the target to be monitored by combining multiple dimensions such as the actual number of turning over, the actual turning over frequency, and the actual intensity score for judgment; the multi-dimensional comprehensive evaluation method can capture subtle changes better than a single indicator, thereby improving the accuracy and reliability of the evaluation; it can automatically integrate the classification results of the classification sub-units, and make judgments based on preset conditions, thereby realizing the automation and intelligence of the evaluation process; it reduces the burden of manual evaluation, reduces human errors, and improves evaluation efficiency; by making staged judgments based on the individual's actual turning over frequency and intensity score, it can provide each person with personalized evaluation results, which helps to more accurately understand the sleep status of the target to be monitored and provide strong support for subsequent interventions.

[0121] Specifically, the posture analysis unit in this embodiment includes: an optimization subunit and a screening subunit;

[0122] The optimization subunit determines an optimization compensation value according to the target position and posture, and optimizes the staged evaluation base value according to the optimization compensation value to obtain a staged evaluation threshold value;

[0123] The stage evaluation threshold = stage evaluation base value + optimization compensation value;

[0124] The screening subunit is used to determine whether the target pressure parameter is abnormal according to the stage evaluation threshold, and determine the target pressure parameter as the key characteristic parameter if it is abnormal;

[0125] Among them, based on the target pressure parameter being greater than or equal to the stage evaluation threshold, the screening subunit determines it as the key feature parameter; based on the target pressure parameter being less than the stage evaluation threshold, the screening subunit determines it as not the key feature parameter.

[0126] Specifically, the optimization subunit in this embodiment analyzes the target position posture through the real-time image acquired by the image acquisition unit to obtain the specific posture of the target to be monitored in the monitoring target area, including: lying sideways, lying flat;

[0127] According to the target position and posture, determine the corresponding optimization compensation value;

[0128] Based on the target position posture being side-lying, the optimization subunit determines a first optimization compensation value;

[0129] Based on the target position posture being lying flat, the optimization subunit determines a second optimization compensation value;

[0130] Applying the optimized compensation value to the stage evaluation base value to obtain a stage evaluation threshold value that is more in line with the actual state of the target;

[0131] The posture analysis unit determines a corresponding staging evaluation base value based on the preliminary staging result;

[0132] Based on the preliminary staging result being a normal stage, the posture analysis unit determines a first staging evaluation base value;

[0133] Based on the preliminary staging result being a mild abnormal stage, the posture analysis unit determines a second staging evaluation base value;

[0134] Based on the preliminary staging result being a severe abnormal stage, the posture analysis unit determines a third staging evaluation base value;

[0135] In this embodiment, the first stage evaluation base value is used to measure the ideal state that the posture pressure of the monitored target should reach when it is in the normal stage; the second stage evaluation base value is used to measure the degree to which the posture pressure of the monitored target deviates from the ideal state when it is in the mild abnormal stage; the third stage evaluation base value is used to measure the degree to which the posture pressure of the monitored target deviates from the ideal state when it is in the severe abnormal stage. The first optimization compensation value is used to measure the adjustment amount of the stage evaluation base value when the target position posture is side-lying in order to more accurately evaluate the target state. The first optimization compensation value takes into account the physiological or physical changes unique to the side-lying posture, such as pressure distribution and muscle force; the second optimization compensation value is used to measure the adjustment amount of the stage evaluation base value when the target position posture is supine; compared with side-lying, the physiological and physical changes in the supine posture are different, so different optimization compensation values ​​are required; the specific setting is affected by individual differences and the monitored target area.

[0136] Specifically, the embodiment of the present invention can adjust the stage evaluation base value according to different target position postures by introducing an optimized compensation value, so as to obtain a stage evaluation threshold that is more in line with the actual state of the target; the personalized adjustment significantly improves the evaluation accuracy of the target posture and pressure parameters; it can adapt to the differences between different individuals, such as weight, body shape, and muscle tension, and reflect the impact of these differences on posture and pressure distribution by adjusting the optimized compensation value; enhance the adaptability and practicality in different users and application scenarios; monitor the target pressure parameters in real time according to the stage evaluation threshold, and once an abnormality is found, it is determined as a key characteristic parameter; it helps to timely discover potential health problems or risks and provide support for subsequent interventions; through the collaborative work of the optimization subunit and the screening subunit, the sleep state of the target to be monitored can be more accurately evaluated.

[0137] Specifically, the driving control module 3 includes: an identification subunit and a signal output subunit;

[0138] The identification subunit is used to perform pattern recognition on the key feature change curve and determine the actual target pattern according to the recognition result;

[0139] The signal output subunit is used to generate a corresponding control signal according to the actual target mode, and send the control signal to the execution operation module 4;

[0140] The control signal includes: control instructions for adjusting the bed surface inclination angle, adjusting the bed surface hardness, and adjusting the bed surface temperature.

[0141] Specifically, the execution operation module 4 includes: a receiving subunit, an action adjustment subunit and a completion confirmation subunit;

[0142] The receiving subunit is used to receive the control signal from the driving control module 3;

[0143] The action adjustment subunit performs corresponding action adjustments according to the control signal, including: adjusting the bed surface inclination angle, adjusting the bed surface hardness, and adjusting the bed surface temperature;

[0144] The completion confirmation subunit sends a corresponding completion signal after the action adjustment is completed to notify the feedback correction module 5 that the action adjustment is completed.

[0145] The feedback correction module 5 includes: a collection subunit, a secondary analysis subunit and a correction subunit;

[0146] The acquisition subunit acquires the actual posture data of the target to be monitored in real time at intermediate time intervals based on the completion signal;

[0147] The secondary analysis subunit performs secondary analysis based on the actual posture data in combination with the real-time monitoring result to determine whether the action adjustment is qualified;

[0148] The correction subunit performs feedback correction according to the secondary analysis result when the action adjustment is unqualified, so as to optimize the accuracy of subsequent action adjustment.

[0149] Specifically, the drive control module 3 determines the actual target mode according to the analysis results of the key feature change curve, and generates a corresponding control signal to guide the execution operation module 4 to make specific action adjustments; performs pattern recognition on the curve, and analyzes the curve characteristics (such as fluctuation amplitude, frequency, trend, etc.) to determine the current actual target mode; the actual target mode includes the need to adjust the bed inclination angle, adjust the bed hardness or adjust the bed temperature; according to the result of pattern recognition, the signal output subunit generates a corresponding control signal; the control signal includes specific adjustment instructions, such as "adjust the bed inclination angle to 30 degrees", "increase the bed hardness to a medium level", "set the bed temperature to 25 degrees Celsius", etc.; the generated control signal is sent to the execution operation module 4 to trigger the corresponding action adjustment.

[0150] The execution operation module 4 receives the control signal from the drive control module 3, and performs specific action adjustments according to the signal content, such as adjusting the inclination angle, hardness or temperature of the bed surface; receives the control signal from the drive control module 3; analyzes the control signal to determine the type of action and specific parameters to be executed; the action adjustment subunit starts the corresponding actuator to adjust the action according to the analyzed control signal.

[0151] For example, if the control signal requires adjusting the inclination angle of the bed surface, the motion adjustment subunit will control the bed surface lifting mechanism to make the bed surface reach the specified inclination angle.

[0152] Similarly, if the control signal requires adjustment of the bed surface hardness or temperature, the motion adjustment subunit will also adjust the mattress inflation volume or temperature control system accordingly;

[0153] After the action adjustment is completed, the completion confirmation subunit will detect whether the adjustment result meets the requirements of the control signal;

[0154] If the adjustment is successful, a completion signal is sent to notify the feedback correction module that the action adjustment is completed;

[0155] If the adjustment fails or does not achieve the expected effect, a fault alarm may be triggered or the process may be readjusted;

[0156] After the action adjustment is completed, the feedback correction module 5 collects the actual posture data and performs secondary analysis to determine whether the action adjustment is qualified; if the adjustment is unqualified, feedback correction is performed according to the secondary analysis results to optimize the accuracy of subsequent action adjustments;

[0157] Based on the completion signal, the acquisition subunit collects the actual posture data of the target to be monitored in real time at the intermediate time interval; the secondary analysis subunit performs secondary analysis based on the actual posture data collected and combined with the real-time monitoring results; the analysis content includes comparing the difference between the actual posture data and the data before adjustment, evaluating the impact of the motion adjustment on the posture and pressure distribution; judging whether the motion adjustment has achieved the expected effect, that is, whether the posture comfort or health status of the target to be monitored has been effectively improved; if the secondary analysis results show that the motion adjustment is unqualified (such as the posture improvement is not obvious, the pressure distribution is still unreasonable, etc.), the correction subunit will feedback and correct according to the analysis results; the correction measures include regenerating the control signal, adjusting the initial time interval, adjusting the optimization compensation value or adjusting the staged evaluation base value; the corrected control signal will be sent to the execution operation module 4 again for a new round of motion adjustment until a satisfactory effect is achieved; through the closed-loop control mechanism, the intelligent bed drive system can continuously optimize the motion adjustment strategy to ensure that the target to be monitored is always in the best posture and comfort state.

[0158] The calculation compensation parameters and the calculation adjustment parameters described in the present invention have two functions: one is to balance the left and right dimensions of the formula, and the other is to adjust the numerical results. No specific assignment is performed in this embodiment. In addition, the calculation formulas in this embodiment are used to intuitively reflect the adjustment relationship between the numerical values, such as positive correlation and negative correlation. Unless otherwise specified, the parameter values ​​that are not specifically limited to numerical values ​​are all positive.

[0159] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0160] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent bed driving system based on human posture recognition, characterized in that: include: A data acquisition module is used to monitor the target in real time to obtain real-time posture data, and collect the target posture data in real time at a preset initial time interval; A data processing module, which is used to determine the preliminary posture staging result of the target to be monitored according to the real-time monitoring result, analyze the target posture data based on the preliminary staging result, determine the key characteristic parameters of the target to be monitored according to the primary analysis result, and generate a corresponding key characteristic change curve according to the key characteristic parameters; A drive control module, which is used to identify the key feature change curve to determine the actual fluctuation situation, determine the actual target mode according to the actual fluctuation situation, and issue a corresponding control signal based on the actual target mode; An execution operation module, which is used to receive the control signal, perform corresponding action adjustment according to the control signal, and send a corresponding completion signal when the action adjustment is completed; A feedback correction module, which collects the actual posture data of the target to be monitored in real time at intermediate time intervals based on the completion signal, performs secondary analysis based on the actual posture data combined with the real-time monitoring result, determines whether the action adjustment is qualified based on the secondary analysis result, and starts the feedback correction mode when the action adjustment is unqualified; The real-time posture data includes: real-time turning over action and real-time action amplitude; The target posture data includes: target position posture and target pressure parameters; The intermediate time interval is equal to 1 / 3 of the initial time interval; The data acquisition module includes: a video monitoring unit, a timing unit, an image acquisition unit and a plurality of pressure sensors; the video monitoring unit is used to monitor the target to be monitored in real time to obtain the real-time turning over action and real-time action amplitude of the target to be monitored; the timing unit starts timing when the actual stay time of the target to be monitored in the monitoring target area meets the timing start condition to obtain the actual monitoring time; the image acquisition unit acquires real-time images every time when the actual monitoring time reaches the initial time interval to obtain the target position and posture; the pressure sensors are evenly distributed in the monitoring target area, and collect the target pressure parameters of the target to be monitored in real time at the initial time interval; The data processing module includes: a determination unit, a posture analysis unit and a generation unit; the determination unit is used to calculate the actual turning frequency according to the real-time turning action and the actual monitoring time, classify and mark the real-time action amplitude, and determine the preliminary staging result according to the actual turning frequency and the classification mark result; the posture analysis unit determines the corresponding staging evaluation base value based on the preliminary staging result, and screens the target pressure parameter according to the staging evaluation base value combined with the target position posture to obtain the key feature parameter; the generation unit is used to generate the corresponding key feature change curve according to the key feature parameter; Wherein, when the actual stay time is greater than or equal to the preset standard stay time, the timing unit determines that the timing start condition is met; when the actual stay time is less than the standard stay time, the timing unit determines that the timing start condition is not met; The determination unit includes: an accumulation subunit, a calculation subunit, a classification subunit, a marking subunit and a judgment subunit; The accumulating subunit is used to accumulate the real-time turning over actions to obtain the actual turning over times, and obtain the actual monitoring time when the actual turning over times meets the calculation start condition; The calculation subunit is used to calculate the actual turning over frequency according to the actual number of turning over times and the actual monitoring time; The classification subunit is used to compare the real-time motion amplitude with a preset standard motion amplitude to determine the record type of the real-time motion amplitude; The marking subunit is used to mark corresponding actions according to the classification result of the classification subunit; The determination subunit, when the actual number of turning over meets the calculation start condition, integrates the classification result of the classification subunit, determines the actual turning over frequency and the number of items in the integration result that meet the single determination condition, and determines the preliminary staging result according to the number of items; when the actual number of turning over is greater than or equal to the preset standard number of turning over, the accumulation subunit determines that the calculation start condition is met; when the actual number of turning over is less than the standard number of turning over, the accumulation subunit determines that the calculation start condition is not met; Wherein, the classification subunit determines the record type according to the real-time amplitude difference and the preset first evaluation value; the real-time amplitude difference is the absolute value of the difference between the real-time action amplitude and the standard action amplitude; When the real-time amplitude difference is less than or equal to the first evaluation value, the classification subunit determines that the record type is a low-intensity action; When the real-time amplitude difference is greater than the first evaluation value and the real-time action amplitude is greater than the standard action amplitude, the classification subunit determines that the record type is a high-intensity action; When the real-time amplitude difference is greater than the first evaluation value and the real-time action amplitude is less than the standard action amplitude, the classification subunit determines that the record type is a normal micro-motion; The determination subunit calculates an actual strength score according to the integration result, and determines whether the integration situation meets the single determination condition according to the actual strength score and a preset standard strength score; Wherein, based on the fact that the actual strength score is greater than or equal to the standard strength score, the determination subunit determines that the integration situation meets the single determination condition; based on the fact that the actual strength score is less than the standard strength score, the determination subunit determines that the integration situation does not meet the single determination condition; The determination subunit determines the preliminary staging result according to the number of items; If the number of items is 0, the determination subunit determines that the preliminary staging result is a normal stage; If the number of items is 1, the determination subunit determines that the preliminary staging result is a mild abnormal stage; If the number of items is 2, the determination subunit determines that the preliminary staging result is a severe abnormal stage; The determination subunit integrates the classification results of the classification subunit; Integrate the real-time movement amplitudes of the low-intensity movements, to obtain a first set; Integrate the real-time action amplitudes of the high-intensity actions recorded in the type to obtain a second set; Read a first quantity in the first set, and read a second quantity in the second set; The determination subunit calculates the actual strength score according to the first number and the second number; The actual strength score=the first quantity×a1+the second quantity×a2; Wherein, a1 is a first calculation compensation parameter for the actual strength score of the first quantity; a2 is a second calculation compensation parameter for the actual strength score of the second quantity.

2. The intelligent bed driving system based on human posture recognition according to claim 1, characterized in that: The posture analysis unit includes: an optimization subunit and a screening subunit; The optimization subunit determines an optimization compensation value according to the target position and posture, and optimizes the staged evaluation base value according to the optimization compensation value to obtain a staged evaluation threshold value; The stage evaluation threshold = stage evaluation base value + optimization compensation value; The screening subunit is used to determine whether the target pressure parameter is abnormal according to the stage evaluation threshold, and determine the target pressure parameter as the key characteristic parameter if it is abnormal; Among them, based on the target pressure parameter being greater than or equal to the stage evaluation threshold, the screening subunit determines it as the key feature parameter; based on the target pressure parameter being less than the stage evaluation threshold, the screening subunit determines it as not the key feature parameter.

3. The intelligent bed driving system based on human posture recognition according to claim 2 is characterized in that: The optimization subunit analyzes the target position and posture through the real-time image acquired by the image acquisition unit to obtain the specific posture of the target to be monitored in the monitoring target area, including: lying sideways and lying flat; According to the target position and posture, determine the corresponding optimization compensation value; Based on the target position posture being side-lying, the optimization subunit determines a first optimization compensation value; Based on the target position posture being lying flat, the optimization subunit determines a second optimization compensation value; The optimized compensation value is applied to the stage evaluation base value to obtain a stage evaluation threshold that is more in line with the actual state of the target.

4. The intelligent bed driving system based on human posture recognition according to claim 3 is characterized in that: The posture analysis unit determines a corresponding staging evaluation base value based on the preliminary staging result; Based on the preliminary staging result being a normal stage, the posture analysis unit determines a first staging evaluation base value; Based on the preliminary staging result being a mild abnormal stage, the posture analysis unit determines a second staging evaluation base value; Based on the preliminary staging result being a severe abnormality stage, the posture analysis unit determines a third staging evaluation base value.

Images