Turning-over monitoring device and turning-over monitoring method

By combining a rolling detection module and an image information acquisition module with a rolling recognition model, the problem of the inability to automatically and accurately determine an infant's rolling over in existing technologies is solved. This enables automatic monitoring and timely notification of an infant's rolling over, improving monitoring efficiency.

CN112315426BActive Publication Date: 2024-11-26WUHAN SIMSHINE INTELLIGENT TECH CO LTD
View PDF 15 Cites 0 Cited by

Patent Information

Application Number
CN202011059706.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-09-30
Publication Date
2024-11-26
Estimated Expiration
2040-09-30

AI Technical Summary

Technical Problem

Existing infant care devices cannot automatically and accurately determine whether the person being cared for has turned over, making it impossible for caregivers to know in a timely manner about the infant's turning over movements during sleep.

Method used

The system employs a turning detection module and an image information acquisition module. By monitoring changes in the pressure area of ​​the bedding and changes in image posture, combined with a turning recognition model, it determines whether the person being cared for has turned over, and outputs monitoring information through the control module.

Benefits of technology

It enables automatic monitoring and timely notification of infants' rolling over, reducing the need for caregivers' focus and improving the accuracy and timeliness of monitoring rolling over movements.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN112315426B_ABST
    Figure CN112315426B_ABST
Patent Text Reader

Abstract

The present invention belongs to the field of electronic monitoring technology, and in particular, relates to a turning-over monitoring device and a turning-over monitoring method. The turning-over monitoring device comprises: a control module, a turning-over detection module and / or an image information acquisition module; the turning-over detection module is used to monitor whether the ward has a turning-over action, and generates a turning-over signal when there is a turning-over action; the image information acquisition module is used to acquire the image information of the ward; the control module is used to judge whether the ward has turned over according to the turning-over signal and / or the image information. The turning-over monitoring method comprises the following steps: S1: monitoring whether the ward has a turning-over action, and generating a turning-over signal when there is a turning-over action; S2: and / or acquiring the image information of the ward; S3: judging whether the ward has turned over according to the turning-over signal and / or the image information. The present invention can automatically and accurately judge whether the ward has turned over.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of electronic monitoring, and in particular to a turning-over monitoring device and a turning-over monitoring method. Background Art

[0002] Sleep is one of the most important activities in people's daily lives. Infants and young children spend more time sleeping. Since they are weak and lack the ability to protect themselves, turning over in sleep may bring danger. Monitors are needed to help care for infants and young children so that guardians can promptly understand the infants' movements during sleep when the infants are out of sight. Currently, most infant care devices have simple image capture functions, which cannot promptly identify and actively notify guardians when infants and young children turn over. Summary of the invention

[0003] In view of this, the embodiments of the present invention provide a turning-over monitoring device and a turning-over monitoring method to solve the technical problem in the prior art that it is impossible to automatically and accurately determine whether a care recipient has turned over.

[0004] The technical solution adopted by the present invention is:

[0005] In a first aspect, the present invention provides a turning-over monitoring device, the device comprising:

[0006] A control module, a turning over detection module and / or an image information acquisition module;

[0007] The turning over detection module is used to monitor whether the ward has a turning over action, and generate a turning over signal when there is a turning over action;

[0008] The image information acquisition module is used to acquire image information of the guardian;

[0009] The control module is used to determine whether the person being supervised has turned over according to the turning over signal and / or the image information.

[0010] Preferably, the turning over detection module is used to determine whether the person being supervised has turned over according to the change of the pressure area of ​​the bedding in a preset time period.

[0011] Preferably, the turning over detection module further comprises a stress detection submodule, and the stress detection submodule is used to detect the stress state of a detection point on the bedding of the monitored person.

[0012] Preferably, the control module is used to determine whether the person under guardianship has turned over according to the detection result of the turning over recognition model on the image information.

[0013] Preferably, it also includes an output module, which is used to output monitoring information and / or a judgment result of whether the person being monitored has turned over.

[0014] In a second aspect, the present invention further provides a method for monitoring turning over, the method comprising the following steps:

[0015] S1: monitors whether the person being monitored has turned over, and generates a turning signal when there is a turning over movement;

[0016] S2: and / or obtaining image information of the ward;

[0017] S3: Determine whether the person being supervised has turned over according to the turning over signal and / or the image information.

[0018] Preferably, the step S3: judging whether the guardian has turned over according to the turning over signal of the turning over detection module and / or the image information of the image information acquisition module comprises:

[0019] S31: Establishing a turning over recognition model;

[0020] S32: inputting the image information into a turning over recognition model for recognition;

[0021] S33: The turning over recognition model outputs a recognition result of whether the guardian turns over according to the image information.

[0022] Preferably, the step S31: establishing a turning over detection model comprises the following steps:

[0023] S311: Acquire images of the guardian in the first state and the second state as turning-over training samples;

[0024] S312: Obtaining a turn-over recognition model through training according to the turn-over training samples.

[0025] Preferably, the step S311: acquiring images of the ward in the first state and the ward in the second state as turning over training samples comprises:

[0026] S3111: Acquire an image corresponding to the moment when the pressure area on the bedding of the monitored person is the largest as the image in the first state;

[0027] S3121: Acquire an image corresponding to the moment when the pressure area on the bedding of the monitored person is the smallest as the image in the second state.

[0028] Preferably, after S33: the turning over recognition model outputs the recognition result of whether the guardian turns over according to the image information, the method further includes:

[0029] S313: Compare the turning-over signal with the recognition result output by the turning-over recognition model;

[0030] S314: Optimize the turning over recognition model according to the comparison result.

[0031] Beneficial effects: The turning-over monitoring device of the present invention utilizes a turning-over detection module to monitor whether the person being cared for has a turning-over action, and generates a turning-over signal when there is a turning-over action. The image information acquisition module is utilized to acquire the image information of the person being cared for, and the control module can automatically determine whether the person being cared for has turned over based on the turning-over signal and the image information of the person being cared for. The turning-over monitoring device of this embodiment can monitor the person being cared for throughout the entire process of the person being cared for during his / her sleep, and automatically determine whether the person being cared for has turned over, so that the user does not need to spend too much time to specifically look after the person being cared for. The turning-over monitoring method of the present invention can monitor the person being cared for throughout the entire process of the person being cared for during his / her sleep, and automatically determine whether the person being cared for has turned over based on the turning-over signal and the image information of the person being cared for, without the need for dedicated care, and can effectively monitor the turning over of the person being cared for. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the drawings required for use in the embodiment of the present invention will be briefly introduced below. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work, and these are all within the protection scope of the present invention.

[0033] Figure 1 1 is a structural block diagram of a turning-over monitoring device in Embodiment 1 of the present invention;

[0034] Figure 2 It is a schematic diagram of the pressure on the bedding when the person being cared for is lying on his side in Embodiment 1 of the present invention;

[0035] Figure 3 It is a schematic diagram of the pressure on the bedding when the person being cared for is lying flat in Example 1 of the present invention;

[0036] Figure 4 It is a structural block diagram of the turning-over monitoring device with an output module and an alarm module in Example 1 of the present invention;

[0037] Figure 5 This is a structural block diagram of a turning-over monitoring device with an intelligent terminal in Embodiment 1 of the present invention;

[0038] Figure 6 This is a flow chart of the turning-over monitoring device in Embodiment 1 of the present invention for determining whether the care recipient has turned over by integrating the turning-over signal of the turning-over detection module and the image information of the image information acquisition module;

[0039] Figure 7 This is a flow chart of the turning over monitoring method in Embodiment 2 of the present invention;

[0040] Figure 8 This is a flow chart of the present invention for determining whether the guardian has turned over;

[0041] Fig. 9 A flow chart of establishing a turning over recognition model for the present invention;

[0042] Fig.10 This is a detailed flow chart of combining the pressure area change and the image information of the cared person to determine the turning over of the cared person in Example 2 of the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution in the embodiment of the present invention will be clearly and completely described in conjunction with the drawings in the embodiment of the present invention. It should be noted that, in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. In the description of the present invention, it should be understood that the orientation or position relationship indicated by the terms "center", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc. is based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. Moreover, the term "include", "comprise" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such a process, method, article or device. In the absence of further restrictions, the elements defined by the phrase "comprising..." do not exclude the presence of other identical elements in the process, method, article or device comprising the elements. If there is no conflict, the various features in the embodiments and examples of the present invention can be combined with each other, all within the scope of protection of the present invention.

[0044] Embodiment 1:

[0045] like Figure 1 As shown, Embodiment 1 of the present invention discloses a turning-over monitoring device, comprising:

[0046] A control module, an output module, a turning over detection module and / or an image information acquisition module;

[0047] The turning over detection module is used to monitor whether the ward has a turning over action, and generate a turning over signal when there is a turning over action;

[0048] The image information acquisition module is used to acquire image information of the guardian;

[0049] The control module is used to determine whether the person being supervised has turned over according to the turning over signal and / or the image information.

[0050] The control module plays the role of overall control in the device. The control module can be a central processing unit (CPU), or other general-purpose processors, single-chip microcomputers, ARM, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc.

[0051] The turning over detection module also includes a plurality of stress detection submodules, and the stress detection submodules are arranged on the bedding (such as a mattress, blanket, quilt, etc.) where the care recipient sleeps. The stress detection submodule may use a stress detection sensor. When the care recipient lies on the bedding, stress will be generated at the place where the care recipient contacts the bedding. In this regard, the present embodiment may set a plurality of detection points at different positions where the care recipient contacts the bedding, and then set a stress detection submodule at the corresponding detection point position, so that when the care recipient lies on the bedding, the stress detection submodule may detect the stress state of the detection point on the bedding of the care recipient. When the stress of the detection point changes, it indicates that the contact state of the care recipient with the detection point also changes. Since the present embodiment adopts the aforementioned structure, the turning over monitoring module can monitor the stress state of each point of the bedding that the care recipient contacts when sleeping. According to the stress state of each detection point on the bedding, the turning over monitoring module can calculate the pressure area of ​​the bedding with all the pressure points of the bedding as the boundary. Figure 2 and Figure 3 As shown in the figure, each ellipse represents a detection point, and a stress sensor is arranged at the position of the point. The ellipse with a cross-section line indicates that the stress at the position exceeds the preset threshold, and the person being cared for exerts pressure on the point. The point is a pressure point, and the corresponding pressure boundary is obtained by enveloping these pressure points with an envelope line. Figure 2 This is a diagram of the pressure state of the bedding when the person being cared for is in a supine position. Figure 2 It can be seen that the pressure area of ​​the bedding is larger at this time. Figure 3 This is a diagram of the pressure state of the bedding when the person being cared for is in a side-lying position. Figure 3 It can be seen that the pressure area of ​​the bedding is smaller at this time.

[0052] The turning over detection module in this embodiment also includes a processor, which is connected to each of the sub-detection modules and can calculate the pressure area of ​​the bedding according to the stress changes detected by the sub-detection modules. The processor of the turning over detection module can be an independent processor or a processor shared with the control module.

[0053] When the turning over monitoring module detects that the change ratio of the pressure area of ​​the bedding within a certain period of time exceeds a certain threshold, it is determined that the care recipient has turned over, and a mode signal indicating that the care recipient has turned over is sent to the control module. The turning over signal may also include the time points corresponding to the maximum and minimum values ​​of the pressure area of ​​the mattress within the set time period. The predetermined time period refers to a certain time length, such as within 20 seconds or within 15 seconds. The specific time length can be determined according to actual conditions and is not limited here.

[0054] Since the pressure area of ​​the bedding will change significantly when the care recipient turns over, and the stress state of the detection point can be quickly and accurately detected by the stress detection submodule, the turning over monitoring module of this embodiment can accurately and timely monitor whether the care recipient has a turning over action.

[0055] In this embodiment, the image information acquisition module can be an electronic device or device with a video acquisition function such as a camera or a camcorder. These electronic devices or devices with a video acquisition function can be arranged around the care recipient. For example, one or more cameras are installed around the care recipient's accessories or surroundings to shoot the care recipient's situation during sleep, and the video or image captured by the image information acquisition module is sent to the control module for processing. The control module can judge whether the care recipient has turned over based on the sleeping posture of the care recipient reflected in the acquired image information, such as supine, side-lying, etc. For example, if it is detected that the care recipient changes from a supine posture to a side-lying posture within a set time period, it can be judged that the care recipient has turned over.

[0056] like Figure 4As shown, the turning over monitoring device of this embodiment can also be provided with an output module and an alarm module. The output module is connected to the control module, and the output module has local communication and remote communication functions. The output module is used to output monitoring information and / or the judgment result of whether the person being cared for turns over. For example, the output module can output the stress state change of each detection point detected by the stress detection module, the pressure position of the bedding detected by the turning over detection module, the size of the pressure area, the time when the turning over signal is generated, the video image of the person being cared for and other monitoring information. When the control module determines that the person being cared for turns over, it can send a trigger signal indicating that the person being cared for turns over to the alarm module, and the alarm module automatically sends the alarm information of the person being cared for turning over to the relevant personnel after receiving the trigger signal, so that the relevant personnel can know the turning over of the person being cared for in time. The alarm module can generate an alarm signal to remind the user by means of sound, light, vibration, etc.

[0057] In addition, if Figure 5 As shown, the turning-over monitoring device of this embodiment also includes an intelligent terminal with communication function, such as a digital mobile phone (mobile phone), a personal computer, a tablet computer, an intelligent wearable device (smart watch, smart bracelet), an intelligent home appliance (smart TV, smart TV box, smart air conditioner, smart refrigerator, smart dressing mirror, smart speaker, smart socket), etc. Among them, the turning-over detection module and / or the image information acquisition module, the control module, the output module, the alarm module and the aforementioned terminal with communication function can communicate with each other by wired or wireless means. The aforementioned terminal can also communicate with each module in the turning-over monitoring device, for example, passively receiving the turning-over information of the caregiver sent by the output module through a mobile phone or a smart watch, and can also actively view the current picture of the camera, and can also receive the alarm signal sent by the alarm module. In this way, the user can be free from distance restrictions and can easily understand the turning-over situation of the caregiver at any time. In order to facilitate user operation, the corresponding software program can be installed in the aforementioned terminal, such as installing an APP in a mobile phone, and the user receives the turning-over information of the caregiver through the corresponding APP program, or views the current picture of the camera.

[0058] Example 2

[0059] The control module of this embodiment can determine whether the caregiver has turned over based on the detection data of the turning over detection module alone, or can determine whether the caregiver has turned over based on the image information obtained by the image information acquisition module alone. The control module can use the turning over recognition model to identify the image information obtained by the image information acquisition module, and identify some specific sleeping postures of the caregiver from the image information, such as lying flat, lying on the side, etc., and then determine whether the caregiver has turned over based on the changes in these sleeping postures. The turning over recognition model can be obtained repeatedly through sample training.

[0060] In addition, the control module of this embodiment can also combine the detection data of the turning over detection module and the image information obtained by the image information acquisition module to comprehensively judge the turning over of the care recipient, so that the judgment of whether the care recipient turns over is more accurate and faster. Figure 6 As shown, the specific steps may be:

[0061] Step S101, monitoring the sleeping status of infants and young children through a turning over monitoring module;

[0062] Step S102, the turning over monitoring module determines whether the infant has a turning over action;

[0063] Step S103, the control module extracts the image at the time point of turning over and stores it as a sample;

[0064] Step S104, using the SVM classifier to train the sample data into a turn-over recognition model;

[0065] Step S105, evaluating the quality of the turn-over recognition model trained in the previous step, if the evaluation result is poor, repeat S103-S105, if the evaluation result is good, proceed to the next step;

[0066] Step S106, if the evaluation result is excellent, a turn-over recognition model is generated;

[0067] Step S107: No turning over action is detected and no response is made.

[0068] Example 3

[0069] like Figure 7 As shown, this embodiment provides a method for monitoring turning over, which includes the following steps:

[0070] S1: monitors whether the person being monitored has turned over, and generates a turning signal when there is a turning over movement;

[0071] The specific monitoring method is: monitoring the changes in the pressure area of ​​the bedding of the monitored person. When the change in the pressure area exceeds a preset threshold, it is considered that the monitored person has turned over, and a turning signal is sent in time.

[0072] The method for calculating the change in the pressure area of ​​the bedding of the care recipient may be to determine the boundary of the contact area between the care recipient and the bedding by detecting the pressure state of the detection point on the bedding, and then calculate the pressure area of ​​the bedding based on the boundary. The detection point may be a position on the bedding that may be in contact with the care recipient.

[0073] S2: and / or obtaining image information of the ward;

[0074] Electronic devices or devices with video acquisition functions such as cameras and video cameras can be placed around the care recipient. For example, one or more cameras can be installed around the care recipient to capture the care recipient's sleep and obtain image information of the care recipient.

[0075] S3: Determine whether the person being supervised has turned over according to the turning over signal and / or the image information.

[0076] After the image information is obtained, it can be determined whether the care recipient has turned over based on the sleeping posture of the care recipient reflected in the image information, such as lying flat, lying on the side, etc. For example, if it is detected that the care recipient changes from lying flat to lying on the side within a set time period, it can be determined that the care recipient has turned over.

[0077] In order to quickly and accurately determine whether the person being cared for has turned over based on the image information, Figure 8 As shown, in this embodiment, S3: judging whether the guardian has turned over according to the turning over signal and / or the image information further includes:

[0078] S31: Establishing a turning over recognition model;

[0079] S32: inputting the image information into a turning over recognition model for recognition;

[0080] S33: The turning over recognition model outputs a recognition result of whether the guardian turns over according to the image information.

[0081] This embodiment determines whether to turn over by establishing a recognition model. Specifically, the acquired image information is input as input to the turning over recognition model, and the turning over recognition model processes the input image information to obtain a recognition result of whether to turn over, and then outputs the recognition result.

[0082] The above-mentioned turning over recognition model can be obtained through repeated sample training. Fig. 9 As shown, the specific steps are:

[0083] S311: Acquire images of the guardian in the first state and the second state as turning-over training samples.

[0084] S312: Obtaining a turn-over recognition model through training according to the turn-over training samples.

[0085] The image in the first state may be an image of the care recipient in a supine state, and the image in the second state may be an image of the care recipient in a side-lying state. The image of the two states is used as samples to train the obtained turn-over recognition model to process the acquired image information, and it can accurately identify whether the care recipient in the image is in a supine or side-lying state. If the care recipient's sleeping posture changes (for example, from a supine state to a side-lying state, or from a side-lying state to a supine state, or from a side-lying state to a flat state, or from a side-lying state to a supine state and then to a side-lying state), it can be considered that the care recipient has turned over.

[0086] Wherein, S311: acquiring images of the guardian in the first state and images of the guardian in the second state as turning over training samples includes:

[0087] S3111: Acquire an image corresponding to the moment when the pressure area on the bedding of the monitored person is the largest as the image in the first state;

[0088] S3121: Acquire an image corresponding to the moment when the pressure area on the bedding of the monitored person is the smallest as the image in the second state.

[0089] This embodiment uses the size of the pressure area on the bedding as the standard for judging whether the care recipient is lying flat or lying on the side, which can accurately reflect the sleeping position of the care recipient. In addition, the size of the pressure area on the bedding is easy to accurately detect and obtain, so it can quickly and accurately judge whether the care recipient has turned over.

[0090] When the above scheme is implemented, when it is monitored that the change ratio of the mattress pressure area exceeds a certain threshold within a certain period of time, it can be judged that the baby has turned over, and the corresponding time points of the maximum and minimum values ​​of the mattress pressure area during this period of time can be obtained. When the turning over signal and the above time point information are obtained, the images corresponding to the time points with the maximum and minimum pressure areas are immediately classified and stored. The images corresponding to the time points with the maximum and minimum pressure areas collected are marked as samples of supine and lateral lying, respectively, and the turning over recognition model is trained accordingly. When training the turning over recognition model, the SVM classifier can be used to train the stored image sample information.

[0091] In order to further improve the accuracy of judging the turning over of the care receiver, the present embodiment may also combine the change of the pressure area with the acquired image information to comprehensively judge the turning over of the care receiver.

[0092] like Fig.10 As shown, the specific method can be:

[0093] Step A0: using a stress detection sensor to detect stress change points when the care recipient is lying down, and calculating the area of ​​the pressure area;

[0094] Step B0: When the stress detection sensor detects a stress change, the pressure area change is calculated;

[0095] Step C0: Determine whether the change in the pressure area exceeds a preset threshold;

[0096] Step D0: Record the time points when the pressure area is the largest and the smallest;

[0097] Step E0: storing the images at the time points with the largest and smallest areas as supine and lateral samples respectively;

[0098] Step F0: Using the SVM classifier to train the sample data of supine and side-lying patients into a turning over recognition model;

[0099] Step G0: Detect the video image using the turning over recognition model;

[0100] Step H0: Compare the detection result of the turn-over recognition model with the judgment result of the turn-over sensor to evaluate the quality of the model. If it is good, proceed to the next step; if it is bad, repeat steps E0-G0;

[0101] Step I0: Complete the training of the turning over recognition model;

[0102] Step J0: If the pressure area changes slightly, it is determined that the person has not turned over, and no response is made.

[0103] The above method cleverly utilizes the detected changes in the pressure area of ​​the care recipient's bedding to assist in classifying the care recipient's sleeping posture in the video image case, so that the images corresponding to different sleeping postures can be accurately classified, stored and marked. Then, these accurately classified images are used to continuously train to obtain a turn-over recognition model, and the detection results of the turn-over recognition model are continuously compared with the judgment results of the turn-over detection module, so that the recognition accuracy of the turn-over recognition model is further improved. On the one hand, this embodiment utilizes the high accuracy of judging whether the care recipient has turned over by the change in the pressure area calculated according to the stress state, and uses the method of training the turn-over recognition model based on a large number of samples classified according to the sleeping posture to make the judgment of the turn-over of the care recipient very accurate. On the other hand, the trained turn-over recognition model only needs to recognize the collected video images, making the recognition process simpler and faster.

[0104] In other embodiments, other methods may be used to detect the sleeping posture of the care recipient, such as using an angular velocity or acceleration signal measured by an IMU (inertial measurement unit) to determine whether the care recipient is in a supine or side-lying state. Alternatively, an image of the care recipient in the image information may be obtained, and then the ratio of the height and length of the care recipient's image may be used to determine whether the care recipient is in a supine or side-lying state.

[0105] The above is a detailed introduction to the turning-over monitoring device, apparatus, equipment and storage medium provided in the embodiments of the present invention.

[0106] It should be clear that the present invention is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is 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 the steps after understanding the spirit of the present invention.

[0107] The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present invention are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0108] It should also be noted that the exemplary embodiments mentioned in the present invention 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 steps, that is, the steps can be performed in the order mentioned in the embodiments, or in a different order from the embodiments, or several steps can be performed simultaneously.

[0109] The above is only a specific implementation of the present invention. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the system, module and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. It should be understood that the protection scope of the present invention is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be covered within the protection scope of the present invention.

Claims

1. A turning-over monitoring device, characterized in that: The turning-over monitoring device is used for monitoring the turning-over of infants and young children, and comprises: Control module, turning over detection module and image information acquisition module; The turning over detection module is used to monitor whether the infant has a turning over action, and generates a turning over signal when there is a turning over action; the turning over detection module also includes a stress detection submodule, and the stress detection submodule is used to detect the stress state of the detection point on the infant's bedding; the turning over detection module uses all the pressure points of the bedding as boundaries, calculates the pressure area of ​​the bedding, and arranges stress sensors at the positions of each detection point, wherein, if the stress at the position of the monitoring point with a section line exceeds a preset threshold, the monitoring point is a pressure point, and the pressure boundary is obtained by enveloping the pressure point by making an envelope line; The image information acquisition module is used to acquire image information of infants and young children; The control module is used to determine whether the infant has turned over according to the turning over signal and the image information; The invention provides real-time monitoring of turning over of infants and young children, and determines whether the infants and young children have turned over according to the change of the pressure area calculated by the stress state; when the turning over monitoring module detects that the change ratio of the pressure area of ​​the bedding within a certain period of time exceeds a certain threshold, it is determined that the infants and young children have turned over, and sends a mode signal indicating that the infants and young children have turned over to the control module, and the turning over signal also includes the time points corresponding to the maximum and minimum values ​​of the pressure area of ​​the mattress within the set time period; the time points of the maximum and minimum pressure areas are recorded; the images of the time points of the maximum and minimum areas are stored as supine and side-lying samples respectively; the sample data of supine and side-lying are trained into a turning over recognition model; the turning over recognition model is used to detect video images; the detection result of the turning over recognition model is compared with the judgment result of the turning over sensor in the turning over detection module, and the turning over recognition model is optimized according to the comparison result; the turning over recognition model outputs the recognition result of whether the infants and young children have turned over according to the image information.

2. The turning-over monitoring device according to claim 1, characterized in that: The turning over detection module is used to determine whether the infant has turned over according to the change of the pressure area of ​​the bedding in a preset time period.

3. The turning-over monitoring device according to claim 1, characterized in that: It also includes an output module, which is used to output monitoring information and / or a judgment result of whether the infant has turned over.

4. A method for monitoring turning over, characterized in that: The turning over monitoring method is used for turning over monitoring of infants and young children, and comprises the following steps: S1: monitors whether the infant has a turning movement and generates a turning signal when there is a turning movement; S2: Obtaining image information of infants and young children; S3: judging whether the infant has turned over according to the turning over signal and image information; S3: judging whether the infant has turned over according to the turning over signal of the turning over detection module and / or the image information of the image information acquisition module comprises: S31: generating a turning over recognition model by combining the turning over signal detected by the turning over detection module and the image information acquired by the image information acquisition module to comprehensively judge the turning over of the infant; S32: inputting the image information into a turning over recognition model for recognition; S33: the turning over recognition model outputs a recognition result of whether the infant turns over according to the image information; The turning over detection module uses all the pressure points of the bedding as the boundary, calculates the pressure area of ​​the bedding, and arranges stress sensors at each detection point. If the stress at the monitoring point with the section line exceeds the preset threshold, the monitoring point is the pressure point, and the pressure boundary is obtained by enveloping the pressure point with an envelope line. The step S31: establishing a turning over detection model comprises the following steps: S311: Perform real-time turning-over monitoring on the infant, determine whether the infant turns over according to the change of the pressure area calculated by the stress state, and obtain images of the infant in the first state and the second state as turning-over training samples; S312: training a turn-over recognition model according to the turn-over training sample; The step S311: acquiring an image of the infant in the first state and an image of the infant in the second state as turning over training samples comprises: S3111: Acquire an image corresponding to the moment when the pressure area on the infant bedding is the largest as the image in the first state; S3121: Acquire an image corresponding to the moment when the pressure area on the infant bedding is the smallest as the image in the second state; After S33: the turning over recognition model outputs the recognition result of whether the infant turns over according to the image information, the following further includes: S313: Compare the turning-over signal with the recognition result output by the turning-over recognition model; S314: Optimizing the turning over recognition model according to the comparison result, the turning over recognition model outputs a recognition result of whether the infant has turned over according to the image information.

Citation Information

Patent Citations

  • Intelligent identification method and device for baby sleep position

    CN102789672A

  • Gerocamium intelligent nursing system and method based on Internet of Things technology

    CN103325080A

  • Touch photographing device and method based on pressure sensor, and mobile terminal

    CN105159594A

  • Touch-screen photographing method, device for mobile terminal and mobile terminal

    CN105912158A

  • Nighttime human sleeping posture monitoring method and system based on infrared image

    CN105930778A