CN117806214A

CN117806214A - Intelligent mattress control method, device and equipment based on AI algorithm and reverse S support

Info

Publication number: CN117806214A
Application number: CN202311865193.2A
Authority: CN
Inventors: 吕传林; 何梦君
Original assignee: Shenzhen Zhongke Sensor Technology Co ltd
Current assignee: Shenzhen Zhongke Sensor Technology Co ltd
Priority date: 2023-12-29
Filing date: 2023-12-29
Publication date: 2024-04-02

Abstract

The invention relates to the technical field of intelligent home furnishing, solves the technical problem that an intelligent mattress in the prior art cannot accurately judge the sleeping state of a user, and is difficult to fall asleep and deep sleep, and provides an intelligent mattress control method, device and equipment based on an AI algorithm and reverse S support. The method comprises the following steps: acquiring human body sign data of a target object and real-time video data of a target area in a sleep scene, and decomposing the real-time video data into multi-frame target images; processing each target image by using a preset breath detection algorithm, and determining the breath frequency information of a target object; according to the respiratory frequency information, combining the human body sign data, and judging the real-time sleep state of a target object; and controlling the mattress to switch to a working mode corresponding to the real-time sleep state according to the real-time sleep state of the target object. The invention meets the requirements of users in different sleep states and solves the technical problem that the users are difficult to fall asleep and deep sleep.

Description

Intelligent mattress control method, device and equipment based on AI algorithm and reverse S support

Technical Field

The invention relates to the technical field of intelligent home furnishing, in particular to an intelligent mattress control method, device and equipment based on an AI algorithm and reverse S support.

Background

With the continuous improvement of the quality of life requirements, the comfort requirements of the bed and chair products are further improved, the concepts and markets of private customization are also increased, and the starting point of the private customization of the bed and chair field is not just the appearance requirements that the customer expects to be distinctive, but the essence is that the products purchased by the customer can be changed along with the changes of the height and weight of the customer and the psychological and living habits.

Different sleep states have different effects on sleep quality, deep sleep is critical for recovery and repair of the body, while shallow sleep and awake states may be accompanied by lighter body movements; by accurately judging the sleeping state of the user, the hardness, the temperature or the supporting force of the mattress can be correspondingly adjusted so as to promote deep sleep to the greatest extent, improve the overall sleeping quality and enable different users to have different demands on sleeping environments; some people may require a softer mattress while sleeping and a stiffer support while sleeping deeply; through individualized adjustment, the mattress can satisfy user's unique demand better, provides more comfortable sleep experience. There is a close correlation between respiration and heart rate and sleep states, with respiration and heart rate being generally faster in awake states and different patterns in asleep, light sleep and deep sleep states, and monitoring changes in these physiological signals may provide information about the current sleep state of the user. The prior Chinese patent CN 110731783A discloses a novel peak extraction method for heart rate estimation, which comprises the following steps: the high-precision angle sensor is used for measuring the fine pulsation of the human body caused by the pulsation of the heart of the human body, and the change period of the electric signal of the sensor after the filtering treatment is the heart rate. Installing a high-precision single-axis inclination sensor at the position of the bottom of a common mattress, which is close to the chest of a human body, so as to record the vibration and respiratory motion of the heart, and acquiring BCG data by using the inclination sensor; and detecting the maximum difference value of the local adjacent wave crests and wave troughs, detecting the heartbeat of the BCG signal peak mode corresponding to single heart beat, and estimating the heart rate by multiple times of judgment. However, proper positioning and mounting of the sensors under the mattress is critical to accurately measure heart rate, and inaccurate mounting of BCG sensors can result in data distortion or failure to effectively capture small vibrations of the heart; the body size, weight and sleeping posture of different people may influence the measuring effect of the sensor. Some people may have difficulty accurately capturing heart vibrations due to small body size or sleep posture, and thus may not accurately determine the sleep state of the user.

When the user is in an awake state, the massage function on the existing intelligent mattress can help to relieve muscle tension and fatigue by simulating a manual massage mode, which is very important for relieving muscle fatigue after sitting, standing or exercising for a long time in daily life, and the massage function on the mattress can relieve pressure points by improving body posture, thereby reducing the pressure feeling of the body, providing more comfortable sleep experience, enabling the body to go to sleep more quickly by comfortable massage, and improving the overall sleep quality; meanwhile, comfortable music can be played, the comfortable music can help to relax mind and body, relieve stress and anxiety, create a pleasant environment for falling asleep, and the proper music selection can stabilize heartbeat, adjust breathing, help a user to enter a sleep state and improve overall sleep quality; music also effectively masks noise in the environment, particularly for people living in noisy urban environments, music conditioning helps to provide a more calm sleeping environment, and music conditioning functions on mattresses often allow users to select their favorite music or sounds creating a unique sleeping atmosphere for everyone that meets their personality and taste. However, after the user goes to sleep from a waking state, the required music volume and massage force are different, and the music volume and massage force cannot be adjusted according to the actual requirement of the user in the prior art.

After the user is judged to enter a deep sleep state, the user can be in various sleeping postures, for example, the user is in supine in the previous habit, but recently, the user is changed into lateral lying due to respiratory disorder and other reasons caused by the relaxation of respiratory muscles, if the user is in supine habit in the sleeping process, the thickness of a comfort supporting layer of the mattress to a human body is mainly the supporting height of a waist bow position, if the height is excessively larger than the supporting height, the buttocks are excessively lowered and the legs are reversely tilted, the exaggerated shape is like sleeping in a round bottom pot, the spine is inevitably deformed for a long time, and the lumbar spine is externally protruded; however, if the comfort layer is the thickness of the comfort layer for lying on the back, the hip bone is uncomfortable due to the insufficient thickness of the comfort layer and touches the hard support layer in the mattress, and meanwhile, the waist is also bent downwards due to the insufficient thickness of the comfort layer, which can not fully support the waist side, and the scoliosis deformation can be caused for a long time. Thus, the invariable structural factors of existing mattresses are difficult to meet with varying demands of the user, so that a customer would expect to have a custom-made personal custom product that truly fits him. However, because the proportion of the physical structure of each person is also changed with time, and the invariable structure of the prior mattresses is even the invariable structure of the products customized in a certain state is not expected to meet the real requirement of customers at all. The prior Chinese patent CN111053396A discloses a soft and hard adjustable mattress and an adjusting method thereof, comprising the following steps: the mattress body comprises a supporting plate, contact air bags which are sequentially arranged on the supporting plate in the direction from the head of the bed to the tail of the bed and can sense whether the mattress is flattened or not, and a comfortable layer laid on the contact air bags; the air pressure device comprises an air pump for providing pressure air for the contact air bags, electromagnetic air valves connected with the contact air bags in a one-to-one correspondence manner, air pressure detectors communicated with interfaces of the electromagnetic air valves, and an exhaust electromagnetic valve communicated with the air pressure detectors; the human body physiological detection device is used for detecting whether the human body lying on the comfort layer has body movement or not; and the controller is used for enabling the air pressure device to work when the human body physiological detection device detects body movement so as to determine the positions of different areas of the human body and regulate the pressure of the contact air bags corresponding to the different areas of the human body, so that the supporting force of the contact air bags on the different areas of the human body is uniformly distributed. The above patent designs have some advantages, such as adjusting the support force of the mattress according to the body movements of different areas of the human body, but may also have some drawbacks: for example, when there are other non-human items in the bed, the pressure determination by the contact air bag is affected, thus forming a false determination, resulting in a false adjustment of the air bag.

Therefore, how to accurately judge the sleep state of the user, and switch the intelligent mattress to the working mode corresponding to the sleep state of the user, so as to avoid the problems of difficult sleeping and difficult deep sleeping of the user

Disclosure of Invention

In view of the above, the invention provides an intelligent mattress control method, device and equipment based on an AI algorithm and an inverse S support, which are used for solving the problems that in the prior art, an intelligent mattress cannot accurately judge the sleeping state of a user, and the user is difficult to fall asleep and deep sleep.

The technical scheme adopted by the invention is as follows:

in a first aspect, the present invention provides an intelligent mattress control method based on an AI algorithm and an inverse S support, the method comprising:

s1: acquiring human body sign data of a target object and real-time video data of a target area in a sleep scene, and decomposing the real-time video data into multi-frame target images;

s2: processing each target image by using a preset breath detection algorithm, and determining the breath frequency information of a target object;

s3: according to the respiratory frequency information, combining the human body sign data, and judging the real-time sleep state of a target object;

s4: and controlling the mattress to switch to a working mode corresponding to the real-time sleep state according to the real-time sleep state of the target object.

Preferably, the S2 includes:

s21: establishing Gaussian pyramids corresponding to the target images of the continuous frames according to the target images of the continuous frames, wherein the level of each Gaussian pyramid is preset;

s22: screening the corresponding continuous pixel points of the target image of each frame according to the pixel values of the pixel points in each stage of the Gaussian pyramid to obtain each target pixel point belonging to weak motion;

s23: establishing corresponding Laplacian pyramids of all layers according to all the target pixel points;

s24: obtaining the target Laplacian pyramid image according to the Laplacian pyramid of each layer;

s25: superposing the target Laplacian pyramid image and the target image of each continuous frame to obtain a target video for carrying out data enhancement on weak motion image data;

s26: and obtaining the motion frequency of the weak motion corresponding to the target video as the respiratory frequency information of the target object according to the video duration and the motion times of the weak motion of the target video.

Preferably, the S4 includes:

s41: if the real-time sleep state is a waking state or a preliminary sleep-entering state, switching to a first working mode of sleep-entering and sleep-assisting, wherein the first working mode is used for adjusting the sleep-assisting volume and the massage force according to a preset adjustment level;

S42: if the real-time sleep state is a shallow sleep state, closing the first working mode;

s43: and if the real-time sleep state is a deep sleep state, switching to a second working mode of deep sleep self-adaption, wherein the second working mode is used for adjusting all air bags in the mattress to a target state conforming to the real-time sleep state according to the real-time sleep state of the target object.

Preferably, the first operation mode is implemented by:

acquiring a first regulation level, a second regulation level and a third regulation level which are preset and related to sleeping-aid volume and massage strength;

if the real-time sleep state is an awake state, adjusting the sleep-aiding volume and the massage strength to the first adjustment level;

if the real-time sleep state is a preliminary sleep state, the sleep-aiding volume and the massage force are adjusted to the second adjustment level;

and acquiring a time interval after the target object enters the preliminary sleep state, and adjusting the sleep-aiding volume and the massage strength to the third adjustment level if the time interval is larger than a preset duration threshold.

Preferably, the second operation mode is implemented by the following steps:

performing human body key point detection on the target image of each frame by utilizing a pre-trained bone detection algorithm based on deep learning to obtain real-time key point information corresponding to a target object;

Matching the real-time key point information with standard key point information corresponding to each preset sleeping gesture, and outputting a matching result;

acquiring pressure data corresponding to each preset part of a target object according to the real-time key point information;

according to the matching result, combining pressure data corresponding to each preset position to determine the real-time sleeping position of the target object, wherein the real-time sleeping position comprises supine, lateral lying and prone lying;

and adjusting each air bag in the mattress to a target state conforming to the real-time sleeping posture according to the real-time sleeping posture of the target object.

Preferably, the determining the real-time sleeping posture of the target object according to the matching result and combining the pressure data corresponding to each preset position includes:

obtaining the matching result, wherein the matching result comprises: the real-time key point information is matched with one of the supine sleeping posture, the prone sleeping posture and the lateral sleeping posture;

determining the real-time contact area between each preset part and the mattress according to the pressure data corresponding to each preset part;

acquiring a preset area threshold, and if the real-time contact area is larger than the area threshold and the real-time key point information is matched with the supine sleeping position or the prone sleeping position, determining that the real-time sleeping position of the target object is supine or prone;

And if the real-time contact area is smaller than or equal to the area threshold value and the real-time key point information is matched with the side sleeping posture, determining that the real-time sleeping posture of the target object is side sleeping.

Preferably, the adjusting each air bag in the mattress to a target state conforming to the real-time sleeping posture according to the real-time sleeping posture of the target object includes:

and determining a target key point position of a target object according to the real-time key point information, wherein the target key point comprises: head, neck, left side of waist, center of waist, right side of waist, left hip, right hip, left thigh and right thigh;

if the real-time sleeping posture is supine or prone, self-adaptive adjustment is performed on the air bag of the area where the first target key point is located, and the air bag is adjusted to a target state conforming to the real-time sleeping posture, wherein the first target key point comprises the centers of the head, the neck and the waist;

and if the real-time sleeping posture is lateral lying, carrying out self-adaptive adjustment on the air bags of the area where the second target key points are located to be in accordance with the target state of the real-time sleeping posture, wherein the second target key points comprise the head, the neck, the left side of the waist, the left buttocks and the left thigh, or the second target key points comprise the head, the neck, the right side of the waist, the right buttocks and the right thigh.

In a second aspect, the present invention provides an intelligent mattress control device based on AI algorithm and inverted S support, the device comprising:

the data acquisition module is used for acquiring human body sign data of a target object and real-time video data of a target area in a sleep scene and decomposing the real-time video data into multi-frame target images;

the breath detection module is used for processing each target image by utilizing a preset breath detection algorithm and determining the breath frequency information of a target object;

the sleep state analysis module is used for judging the real-time sleep state of the target object according to the respiratory frequency information and the human body sign data;

and the mode switching module is used for controlling the mattress to switch to a working mode corresponding to the real-time sleep state according to the real-time sleep state of the target object.

In a third aspect, an embodiment of the present invention further provides an electronic device, including: at least one processor, at least one memory and computer program instructions stored in the memory, which when executed by the processor, implement the method as in the first aspect of the embodiments described above.

In a fourth aspect, embodiments of the present invention also provide a storage medium having stored thereon computer program instructions which, when executed by a processor, implement a method as in the first aspect of the embodiments described above.

In summary, the beneficial effects of the invention are as follows:

the invention provides an intelligent mattress control method, device and equipment based on an AI algorithm and an inverse S support, wherein the method comprises the following steps: acquiring human body sign data of a target object and real-time video data of a target area in a sleep scene, and decomposing the real-time video data into multi-frame target images; processing each target image by using a preset breath detection algorithm, and determining the breath frequency information of a target object; according to the respiratory frequency information, combining the human body sign data, and judging the real-time sleep state of a target object; and controlling the mattress to switch to a working mode corresponding to the real-time sleep state according to the real-time sleep state of the target object. According to the invention, the human body physical sign data and the real-time video data are acquired, the real-time video data are decomposed into multi-frame target images, and the breathing frequency information and the human body physical sign data are combined, so that the real-time sleep state of a user is judged by fully utilizing various information sources, the method is helpful for accurately judging whether the user is in specific states such as difficult sleeping, difficult deep sleeping and the like, and powerful support is provided for formulating corresponding mattress working modes; meanwhile, according to the real-time sleep state of the target object, the scheme adjusts the mattress to a corresponding working mode, for example, when the user is detected to be in a waking state, the mattress is adjusted to provide gentle massage or gentle vibration so as to help the user relax the body and promote falling asleep; when detecting that the user is in the deep sleep state, the mattress can be adjusted to provide more stable support, the hardness is adjusted to ensure that the user obtains better comfort in the deep sleep, the personalized intelligent adjustment is helpful for improving the sleep quality of the user, and the requirements of the user in different sleep states are met, so that the problems of difficult sleeping and deep sleeping of the user are solved.

Drawings

In order to more clearly illustrate the technical solution of the embodiments of the present invention, the drawings required to be used in the embodiments of the present invention will be briefly described, and it is within the scope of the present invention to obtain other drawings according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flow chart of the overall operation of the intelligent mattress control method based on AI algorithm and reverse S support in embodiment 1 of the invention;

fig. 2 is a schematic flow chart of processing each target image by using a preset breath detection algorithm in embodiment 1 of the present invention;

FIG. 3 is a flow chart of the method for controlling the mattress to switch to the operation mode corresponding to the real-time sleep state in the embodiment 1 of the invention;

FIG. 4 is a flow chart showing the working steps of the first working mode in the embodiment 1 of the present invention;

FIG. 5 is a flow chart showing the working steps of the second working mode in the embodiment 1 of the present invention;

FIG. 6 is a flow chart of determining the real-time sleeping posture of the target object in embodiment 1 of the present invention;

FIG. 7 is a flow chart of adjusting each air bag in the mattress to a target state according to the real-time sleeping posture in embodiment 1 of the present invention;

FIG. 8 is a block diagram of the intelligent mattress control device based on AI algorithm and reverse S support in embodiment 2 of the invention;

fig. 9 is a schematic structural diagram of an electronic device in embodiment 3 of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention more clear, the technical solutions of 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. It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. In the description of the present invention, it should be understood that the terms "center," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," and the like indicate orientations or positional relationships based on the orientation or positional relationships shown in the drawings, merely to facilitate description of the present application and simplify the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the present invention. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element. If not conflicting, the embodiments of the present invention and the features of the embodiments may be combined with each other, which are all within the protection scope of the present invention.

Example 1

Referring to fig. 1, embodiment 1 of the present invention discloses an intelligent mattress control method based on an AI algorithm and an inverse S support, wherein the inverse S support means that the mattress can perform intelligent support according to the body requirement of a target object in a supine or prone position, so as to avoid bending of a spine curve of a user to present an S shape, and realize inverse S support for the user, specifically, an intelligent technology is adopted in mattress design, and by actively matching a sleeping state of the user and combining an intelligent algorithm with a sensor unit and an air pressure execution unit, a sleeping state index and a sleeping posture change of the user are monitored in real time, so as to provide a mattress design with optimal support and pressure distribution, and the intelligent mattress comprises: the mattress comprises a mattress main body, a sensor unit, an execution unit and a control unit; the sensor unit includes: pressure sensor, temperature and humidity sensor and camera; the execution unit includes: the device comprises an air bag adjusting assembly, a temperature and humidity adjusting assembly and a music adjusting assembly; the control unit includes: a power supply, an MCU, a control circuit and a signal processing component, the method comprising:

In the process of acquiring human body sign data of a target object and real-time video data of a target area in a sleep scene, firstly, acquiring heart rate and body movement data of a user, which are acquired by a sign sensor, wherein the sign sensor comprises a heart rate monitor and an accelerometer, and comprehensively and accurately monitoring the sign data of the user; the target object covers people of all ages including children, adults and elders; because the sleeping scene is usually under dark condition, the thermal infrared camera is adopted to collect video data, the real-time video data collected by the high-resolution thermal infrared camera is obtained, the thermal infrared camera is arranged at a position which can fully cover a target area, such as a ceiling and a head of a bed, so that clear and accurate shooting of the target area is ensured, the target area comprises a bed surface area, and the real-time video data is decomposed into multiple frames of target images. By comprehensively utilizing the information of the physical sign sensor and the real-time video data, the physiological and behavioral characteristics of the target object in sleep are comprehensively known, and the comprehensive monitoring method is helpful for more accurately evaluating sleep quality and detecting potential problems and provides customized sleep support for users of different age groups.

in one embodiment, referring to fig. 2, the step S2 includes:

specifically, data enhancement is performed on the target image of each continuous frame, specifically, the position offset distance of each pixel point of the image area caused by respiratory motion is amplified, so that a Laplacian pyramid image corresponding to the image area of respiratory motion is established, the Laplacian pyramid images are different in hierarchy and have different spatial frequencies and signal to noise ratios, the fewer the hierarchy is, the lower the spatial frequency is, and the established Laplacian pyramid image is preferably 3-6 layers.

specifically, the pixel value of each pixel point in the Gaussian pyramid image is recorded as a first pixel value, and the pixel value of each pixel point in the original image is recorded as a second pixel value; and comparing each first pixel value with each second pixel value to obtain a pixel value difference value between any first pixel value and each second pixel value, comparing each pixel value difference value with a pixel value threshold, recording a second pixel value corresponding to a pixel value difference value smaller than the pixel value threshold as a target pixel value, and screening out all target pixel points by the method, wherein the target pixel points do not directly carry out Laplace change to establish a Laplace pyramid image, and the Laplace pyramid image is generated by directly carrying out Laplace change because the target pixel points can be uniformly changed, namely, the gradient of the region disappears after the Laplace change, so that the accuracy of breath detection is affected. The input signal is decomposed into wall edges, textures and smooth components, the strong edges are all pixel points representing the salient pixel values of the whole outline of the image, the textures are all pixel points with small pixel value differences corresponding to all pixel points in the strong edges, namely all pixel points representing the details of the image area corresponding to all pixel points in the strong edges, the smooth components are used for carrying out low-frequency component enhancement and high-frequency component weakening on the image to realize the smooth processing of the image, the Laplace needs to carry out derivation twice, if the Laplace transformation is directly carried out, the processing result of the Laplace transformation is 0 in the uniform color change or gradual change area, and the processed image of the areas becomes holes or disappears, so that the accuracy of respiratory monitoring is affected.

specifically, after each target pixel point is obtained, a Gaussian function is utilized Adjusting each target pixel point to enable the pixel value of each pixel point in the weak motion image to show non-uniform change or non-gradual change, avoiding the existence of holes or gradient disappearance of weak motion in the established Laplace pyramid image, taking-2, -1, 2 and 4 for f respectively to enable the pixel value conversion of a weak motion area to show non-gradual change or uniform change, thereby avoiding zeroing the pixel value of the area after the Laplace conversion, leading the image of the area to disappear, generating a new image after one layer of Gaussian pyramid image is completed, converting the new image into a corresponding layer of Laplace pyramid image, repeating the operation, and finally obtaining the target Laplace pyramid image; by the method, the phenomenon that the image corresponding to the respiratory motion appears holes or disappears due to the Laplace change can be avoided, and the integrity of the respiratory motion data is ensured.

S24: obtaining the target Laplacian pyramid image according to the Laplacian pyramid of each layer; specifically, obtaining Laplacian pyramids of all layers; using the formula Amplifying and superposing the Laplacian pyramid of each layer to obtain the Laplacian pyramidThe target Laplacian pyramid image of weak motion in the infant sleeping video; wherein I (x, t) is the brightness of the pixel at time t, δ (t) is the displacement distance of the corresponding pixel at time t compared with the previous time, α is the amplification factor, and f (x) is the pixel value of the pixel x. Aiming at the generated new local Laplacian pyramid of each layer, according to the principle that the brightness of the same pixel point at the same time is unchanged, the method has the following formula:

I(x,t)＝f(x+δ(t))

then according to the principle of the constant brightness there are:

I(x,t)＝f(x+(1+α)δ(t))

considering that sleep respiratory motion belongs to low-frequency motion, low-pass filtering is performed in a low frequency band, and therefore, the first-order Taylor series expansion is as follows:

and (3) making:

b (x, t) is an image brightness variation signal corresponding to the corresponding breathing signal when the spatial point position in the video channel is x and the time is t.

The method comprises the following steps of amplifying and superposing:

then, overlapping the amplified Laplacian pyramid image with the original image to obtain pyramid images with different scales, and reconstructing the pyramid images to obtain the required amplified video; by amplifying the respiratory motion, the method is beneficial to increasing the identification speed and accuracy of the respiratory motion, thereby ensuring the accuracy of detection.

Specifically, the weak motions include motions caused by respiration, hand inching, waving of hair parts and the like, wave bands corresponding to the weak motions of the target video are researched to determine motion frequencies corresponding to the weak motions, and particularly, the motion frequencies are obtained according to video duration and motion times of the target video, and the obtained motion frequencies are used as respiration frequency information of a target object.

specifically, according to the respiratory rate information, the human body sign data and the predefined sleep state classification, the sleep states are four states of waking, falling asleep, light sleep and deep sleep, and the indexes are integrated to determine the real-time sleep state of the target object, for example: the respiratory rate of the awake state is higher, the body movement is active, the heart rate is higher, and the example range is the respiratory rate: 15-25 break/min, body movement: more frequent exercise, heart rate: 80-100bpm; sleeping state: respiratory rate gradually slows down, body movement gradually decreases, heart rate gradually decreases, example range: respiratory rate: 10-18 break/min, body movement: gradually decrease, heart rate: 60-80bpm; the respiration rate and heart rate of the light sleep state are relatively stable, with little body movement, example ranges: respiratory rate: 8-15breaths/min, body movement: comparatively stationary, heart rate: 50-70bpm, the respiratory rate in the deep sleep state is stable, the body movement is very little, the heart rate is the lowest, and the example range is as follows: respiratory rate: 8-12 break/min, body movement: almost stationary, heart rate: 40-60bpm, and comparing the information with the range by monitoring the respiratory rate, the body movement and the heart rate of the target object in real time and judging the current sleep state of the target object by combining with the rules defined previously. This comprehensive decision helps to more accurately assess sleep state, thereby enabling personalized adjustments to the intelligent mattress and providing more efficient sleep support, it being noted that these ranges and features are general examples, and that specific thresholds and rules need to be optimized for specific situations and individual differences.

Specifically, according to the judgment result of the real-time sleep state of the target object, the working mode of the mattress is adjusted, and the mattress is controlled to be switched to the working mode corresponding to the real-time sleep state. Through such intelligent adjustment, the mattress can better satisfy the demands of users in different sleep states, and personalized sleep support is provided, so that the sleep experience of the users is improved, and meanwhile, the personalized adjustment is hopeful to promote deeper rest and better sleep quality.

In one embodiment, referring to fig. 3, the step S4 includes:

if the real-time sleep state is a waking state or a preliminary sleep-entering state, switching to a first working mode of sleep-entering and sleep-assisting, wherein the first working mode is used for adjusting the sleep-assisting volume and the massage force according to a preset adjustment level;

specifically, if the real-time sleep state is determined to be an awake state, a series of soothing functions are realized by controlling the intelligent mattress to help the target subject to relax his body and promote sleep, including: starting a music playing system, and selecting to play sleep-aiding songs in a preset library, wherein the sleep-aiding songs comprise soft and comfortable music which accords with the preference of a target object, and are beneficial to relaxing the brain and the body; meanwhile, the air bags in the mattress are inflated and deflated to massage the body of the target object, and the massage intensity, frequency and position are adjusted to meet the personalized requirements of the target object, so that the target object is helped to fall asleep quickly.

In an embodiment, please refer to fig. 4, the first operation mode is implemented by:

specifically, three preset adjustment levels are obtained and are respectively used for controlling the sleep-aiding volume and the massage strength, and each adjustment level can be set by a user according to personal preference and comfort level or can be a default standard value of the system. For example: first adjustment level: the device is suitable for the awake state, and provides the maximum volume and the maximum massage force to help the user to relax better in the awake state, so that the user falls asleep quickly; second adjustment level: the device is suitable for a preliminary sleep state, provides weaker volume and massage force, and is beneficial to guiding a user to enter deeper sleep; third adjustment level: is suitable for a period of time after preliminary falling asleep, and provides the lightest volume and massage force to cope with some challenges that users may present during falling asleep, such as slightly waking up.

Specifically, in the awake state, the system adjusts the sleep-aiding volume and the massage force to a first preset adjustment level according to the sleep state information monitored in real time, which means that the system provides the maximum volume and massage force to help the target subject to relax better in the awake state, thereby falling asleep quickly.

specifically, in the preliminary sleep state, the system adjusts the sleep-aiding volume and the massage strength to a preset second adjustment level according to the sleep state information monitored in real time, and the level may provide weaker volume and massage strength, so as to help guide the target object to enter deeper sleep.

Specifically, the system monitors the time after the target object enters the preliminary sleep state, and if the time interval is greater than a preset duration threshold, the system adjusts the sleep-aiding volume and the massage strength to a third adjustment level. This level may provide the lightest volume and massage effort to accommodate some of the demands that the user may have during sleep, such as a slight wake-up. Such dynamic adjustment helps to individually adapt to the needs of the user in different sleep states.

Specifically, if the real-time sleeping state is a deep sleeping state, each air bag in the mattress is adjusted according to the real-time sleeping state of the target object, and the mattress can better adapt to different sleeping states of the user in the deep sleeping state by adjusting the air bags to the target state according with the real-time sleeping state, so that more comfortable support according with a physiological curve is provided. This helps to improve the sleep quality and overall sleep experience of the user.

In an embodiment, please refer to fig. 5, the second operation mode is implemented by:

specifically, a pre-trained deep learning-based skeleton detection algorithm is utilized to detect key points of human bodies on the target images of all frames, the skeleton detection algorithm is completed through the pre-trained deep learning-based skeleton detection algorithm, specifically, VGG pre-train network is used as a backbone network, the network can extract the characteristics of the target images of all frames input, VGG pre-train network is used as the backbone network, and the characteristics of the target images of all frames input are extracted through operations such as rolling and pooling, and play a key role in subsequent skeleton detection; constructing two branch networks, wherein one branch is used for returning the position (L (p)) of a key point, the other branch is used for generating Part Affinity Fields (PAFs, limb relation), the two branches are respectively responsible for capturing the key point position and the limb relation of a target object, in each training state, the network adjusts parameters by calculating a loss function to adapt to the marked key point position and the limb relation, the loss function comprises L2 norm loss of L (p) and S (p), L (p) represents the key point position, S (p) represents the limb relation, and the training process is carried out in each state until the deep learning network converges; after training, the deep learning network can generate a predicted position of each key point in the image and a prediction of the limb relation, which are gradually optimized and generated in each state iterative process, PAFs are used for representing the direction and the strength of the limb, the deep learning network constructs Fields by searching vector directions between pixels connecting each pair of key points, matches adjacent nodes, such as a left wrist node and a left elbow node, calculates the side weight by using PAF of a forearm, finally obtains human body posture information of the whole target object, establishes a skeleton structure of the human body, and identifies the connection and the relative position between the key points as real-time key point information corresponding to the target object.

specifically, acquiring real-time human body key point information from a bone detection algorithm, wherein the key points comprise coordinates of various parts of the body, such as a head, a shoulder, a wrist, a knee and the like;

presetting various sleeping postures, including supine, prone and lateral sleeping, defining a standard key point position in advance for each preset sleeping posture, wherein the standard key point information comprises preset coordinates for each preset sleeping posture;

comparing the real-time key point information with the standard key point information corresponding to each preset sleeping gesture, and calculating the distance or similarity between the real-time key point coordinates and the preset standard coordinates;

determining the current sleeping pose of the target object according to a matching result, wherein the matching result can be a similarity score of each preset sleeping pose;

setting a similarity threshold to determine when the real-time key point information is considered to be successfully matched with a certain preset sleeping gesture;

if the similarity is higher than the set threshold, the matching is considered to be successful, and the current real-time sleeping gesture with the highest similarity is selected.

By matching the real-time key point information with the standard key point information corresponding to the preset sleeping gesture, the system can identify the sleeping gesture of the target object in real time, and provide important information for subsequent sleeping state judgment and mattress adjustment.

in particular, an array of pressure sensors is pre-arranged on the mattress, which sensors are capable of sensing the pressure distribution at the surface of the mattress. Each sensor is responsible for monitoring pressure changes in a particular area; for each preset position, establishing a corresponding relation between the key point and a corresponding sensor area on the mattress, for example, the key point 'head' may correspond to a head area on the mattress, the key point 'shoulder' corresponds to a shoulder area on the mattress, and the like; and mapping the real-time key point information to a corresponding sensor area on the mattress by utilizing a preset corresponding relation, so that the specific position of each key point can be determined, and the corresponding pressure sensor can be associated. And acquiring pressure data of each preset part corresponding to the real-time key point information through a pressure sensor on the mattress. These data reflect the pressure distribution of the target object at different locations on the bed.

Specifically, according to the matching result, combining pressure data corresponding to each preset position to determine a real-time sleeping posture of the target object, wherein the real-time sleeping posture comprises supine, lateral lying and prone lying;

by combining the real-time key point information and the pressure data in the target image, the real-time sleeping gesture of the target object can be more accurately determined;

the pressure data can provide the force distribution on the mattress, the real-time key point information provides the specific position of the body part, and the sleeping gesture can be more accurately identified through the combination of the force distribution and the real-time key point information, so that the possible misjudgment of only depending on the pressure data is avoided: for example, other items such as pillows, toys, etc. may be present on the mattress, which may also exert a certain pressure on the mattress, and relying solely on the pressure data for a determination may be disturbed by these additional factors, resulting in erroneous determinations. By combining the image information, the body part of the target object can be identified, and erroneous judgment caused by other articles on the mattress is avoided. The comprehensive determination of the real-time sleeping posture is not only beneficial to the personalized adjustment of the mattress, but also can provide more comprehensive sleeping analysis, understand the sleeping modes and habits of the user under different sleeping postures, and is beneficial to providing more personalized sleeping advice and improvement schemes for the user.

In an embodiment, referring to fig. 6, according to the matching result, determining the real-time sleeping gesture of the target object in combination with the pressure data corresponding to each preset position includes:

In an embodiment, referring to fig. 7, according to the real-time sleeping posture of the target object, adjusting each air bag in the mattress to a target state conforming to the real-time sleeping posture includes:

in particular, if the real-time sleeping position is supine or prone, the air bag on the mattress is adaptively adjusted for the head area, so that the head is properly supported, which is helpful for maintaining the natural position of the skull and reducing the discomfort of the neck and the spine; the neck is a key area which needs to be well supported in the sleeping process, and proper support of the neck is ensured by adjusting the corresponding air bags so as to maintain the normal physiological curve of the cervical vertebra and relieve the neck pressure; for the central region of the lumbar, the system adjusts the air bags on the mattress to provide support for the lumbar, which helps to maintain the natural curvature of the lumbar, relieve lumbar pressure, and thus improve overall sleep comfort; the self-adaptive adjustment means that the hardness or the height of the air bag is dynamically adjusted according to the target key point position detected in real time, through the self-adaptive adjustment, the mattress can carry out intelligent support according to the body requirement of a target object in a supine or prone position, so that the curve bending of the user backbone is avoided, the S-shaped support is realized, the reverse S-shaped support is realized for the user, the intelligent technology is adopted in the mattress design, and the sleeping state index and the sleeping position change of the user are monitored in real time by actively matching the sleeping state of the user and utilizing a sensor unit and an air pressure executing unit to combine an intelligent algorithm, so that the mattress design with optimal support and pressure distribution is provided. Such personalized adjustments help provide a more comfortable and physiological structured sleep experience, thereby improving the user's falling asleep and sleeping quality.

Specifically, if the real-time sleeping posture is a lateral sleeping posture, the positions of the key points of the head, the neck, the left waist, the left hip and the left thigh in the lateral sleeping posture are mapped to the air bags of the corresponding areas, or the positions of the key points of the head, the neck, the right waist, the right hip and the right thigh in the lateral sleeping posture are mapped to the air bags of the corresponding areas, a feedback loop adjusting algorithm, namely a feedback mechanism, is adopted, the mattress is adjusted in real time by sensing feedback or other physiological data of a user so as to provide more comfortable support, the hardness or the height of the air bags of the corresponding areas are calculated through the real-time key point information and the mapping relation, the adjustment of the support, which possibly comprises the inflation or deflation of the air bags so as to provide the support matched with the lateral sleeping posture, is dynamic, and the system can monitor the change of the positions of the head, the neck, the left waist, the left hip and the left thigh, or the head, the right waist, the right hip and the right thigh of the user in the lateral sleeping posture in real time, and make corresponding air bag adjustment when needed. Adaptive adjustment allows the system to better adapt to individual differences and habits of the user, providing personalized support, ensuring that the user can obtain an optimal sleep experience in a lateral position. Through such an adaptive adjustment mechanism, the mattress system can provide more personalized and physiological structural support in a lateral position, helping the user to fall asleep better and improving overall sleep quality.

Example 2

Referring to fig. 8, embodiment 2 of the present invention further provides an intelligent mattress control device based on AI algorithm and reverse S support, the device comprising:

Specifically, the intelligent mattress control device based on the AI algorithm and the reverse S support provided by the embodiment of the invention comprises: the data acquisition module is used for acquiring human body sign data of a target object and real-time video data of a target area in a sleep scene and decomposing the real-time video data into multi-frame target images; the breath detection module is used for processing each target image by utilizing a preset breath detection algorithm and determining the breath frequency information of a target object; the sleep state analysis module is used for judging the real-time sleep state of the target object according to the respiratory frequency information and the human body sign data; and the mode switching module is used for controlling the mattress to switch to a working mode corresponding to the real-time sleep state according to the real-time sleep state of the target object. The device decomposes the real-time video data into a multi-frame target image by acquiring the human body physical sign data and the real-time video data, and combines the respiratory frequency information and the human body physical sign data, so that the real-time sleep state of the user is judged by fully utilizing various information sources, the device is helpful for accurately judging whether the user is in specific states such as difficult sleeping, difficult deep sleeping and the like, and powerful support is provided for formulating corresponding mattress working modes; meanwhile, according to the real-time sleep state of the target object, the scheme adjusts the mattress to a corresponding working mode, for example, when the user is detected to be in a waking state, the mattress is adjusted to provide gentle massage or gentle vibration so as to help the user relax the body and promote falling asleep; when detecting that the user is in the deep sleep state, the mattress can be adjusted to provide more stable support, the hardness is adjusted to ensure that the user obtains better comfort in the deep sleep, the personalized intelligent adjustment is helpful for improving the sleep quality of the user, and the requirements of the user in different sleep states are met, so that the problems of difficult sleeping and deep sleeping of the user are solved.

Example 3

In addition, the intelligent mattress control method based on the AI algorithm and the inverted S support of embodiment 1 of the present invention described in connection with fig. 1 may be implemented by an electronic device. Fig. 9 shows a schematic hardware structure of an electronic device according to embodiment 3 of the present invention.

The electronic device may include a processor and memory storing computer program instructions.

In particular, the processor may comprise a Central Processing Unit (CPU), or an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), or may be configured as one or more integrated circuits that implement embodiments of the present invention.

The memory may include mass storage for data or instructions. By way of example, and not limitation, the memory may comprise a Hard Disk Drive (HDD), floppy Disk Drive, flash memory, optical Disk, magneto-optical Disk, magnetic tape, or universal serial bus (Universal Serial Bus, USB) Drive, or a combination of two or more of the foregoing. The memory may include removable or non-removable (or fixed) media, where appropriate. The memory may be internal or external to the data processing apparatus, where appropriate. In a particular embodiment, the memory is a non-volatile solid state memory. In a particular embodiment, the memory includes Read Only Memory (ROM). The ROM may be mask programmed ROM, programmable ROM (PROM), erasable PROM (EPROM), electrically Erasable PROM (EEPROM), electrically rewritable ROM (EAROM), or flash memory, or a combination of two or more of these, where appropriate.

The processor reads and executes the computer program instructions stored in the memory to implement any of the intelligent mattress control methods based on AI algorithm and anti-S support in the above embodiments.

In one example, the electronic device may also include a communication interface and a bus. The processor, the memory, and the communication interface are connected by a bus and complete communication with each other, as shown in fig. 9.

The communication interface is mainly used for realizing communication among the modules, the devices, the units and/or the equipment in the embodiment of the invention.

The bus includes hardware, software, or both that couple the components of the device to each other. By way of example, and not limitation, the buses may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an infiniband interconnect, a Low Pin Count (LPC) bus, a memory bus, a micro channel architecture (MCa) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a video electronics standards association local (VLB) bus, or other suitable bus, or a combination of two or more of the above. The bus may include one or more buses, where appropriate. Although embodiments of the invention have been described and illustrated with respect to a particular bus, the invention contemplates any suitable bus or interconnect.

Example 4

In addition, in combination with the intelligent mattress control method based on the AI algorithm and the inverted S support in the above embodiment 1, embodiment 4 of the present invention may also provide a computer readable storage medium. The computer readable storage medium has stored thereon computer program instructions; the computer program instructions, when executed by the processor, implement any of the intelligent mattress control methods of the above embodiments based on AI algorithm and anti-S support.

In summary, the embodiment of the invention provides an intelligent mattress control method, device and equipment based on an AI algorithm and reverse S support.

It should be understood that the invention is not limited to the particular arrangements and instrumentality described above and shown in the drawings. For the sake of brevity, a detailed description of known methods is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method processes of the present invention are not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between steps, after appreciating the spirit of the present invention.

The functional blocks shown in the above-described structural block diagrams may be implemented in hardware, software, firmware, or a combination thereof. When implemented in hardware, it may be, for example, an electronic circuit, an Application Specific Integrated Circuit (ASIC), suitable firmware, a plug-in, a function card, or the like. When implemented in software, the elements of the invention are the programs or code segments used to perform the required tasks. The program or code segments may be stored in a machine readable medium or transmitted over transmission media or communication links by a data signal carried in a carrier wave. A "machine-readable medium" may include any medium that can store or transfer information. Examples of machine-readable media include electronic circuitry, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio Frequency (RF) links, and the like. The code segments may be downloaded via computer networks such as the internet, intranets, etc.

It should also be noted that the exemplary embodiments mentioned in this disclosure describe some methods or systems based on a series of steps or devices. However, the present invention is not limited to the order of the above-described steps, that is, the steps may be performed in the order mentioned in the embodiments, or may be performed in a different order from the order in the embodiments, or several steps may be performed simultaneously.

In the foregoing, only the specific embodiments of the present invention are described, and it will be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the systems, modules and units described above may refer to the corresponding processes in the foregoing method embodiments, which are not repeated herein. It should be understood that the scope of the present invention is not limited thereto, and any equivalent modifications or substitutions can be easily made by those skilled in the art within the technical scope of the present invention, and they should be included in the scope of the present invention.

Claims

1. An intelligent mattress control method based on an AI algorithm and an inverse S support, which is characterized by comprising the following steps:

2. The intelligent mattress control method based on AI algorithm and inverse S support of claim 1, wherein S2 comprises:

3. The intelligent mattress control method based on AI algorithm and inverse S support of claim 1, wherein S4 comprises:

4. The intelligent mattress control method based on AI algorithm and reverse S support of claim 3, wherein the first mode of operation is implemented by:

5. The intelligent mattress control method based on AI algorithm and reverse S support of claim 3, wherein the second mode of operation is implemented by:

6. The intelligent mattress control method based on AI algorithm and inverse S support of claim 5, wherein determining the real-time sleeping posture of the target object according to the matching result in combination with pressure data corresponding to each preset part comprises:

7. The intelligent mattress control method based on AI algorithm and anti-S support of claim 5, wherein adjusting each air bag in the mattress to a target state that meets the real-time sleeping posture in accordance with the real-time sleeping posture of the target object comprises:

8. An intelligent mattress control device based on AI algorithm and reverse S support, the device comprising:

9. An electronic device, comprising: at least one processor, at least one memory, and computer program instructions stored in the memory, which when executed by the processor, implement the method of any one of claims 1-7.

10. A storage medium having stored thereon computer program instructions, which when executed by a processor, implement the method of any of claims 1-7.