A mirror-type children's sitting posture monitoring robot
By designing a mirror-type children's sitting posture monitoring robot, using cameras and visual recognition modules to detect the sitting posture status of children, and adjusting the tilt angle of the robot through the angle adjustment mechanism, the problems of low monitoring accuracy and negative impact on health of existing equipment are solved, and efficient and safe sitting posture correction is achieved.
Patent Information
- Application Number
- CN202411223299.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-03
- Publication Date
- 2025-03-04
- Estimated Expiration
- 2044-09-03
AI Technical Summary
Existing children's sitting posture correction equipment has low monitoring accuracy, frequent false reminders and negative effects on physical health, and passive correction equipment may lead to bone deformation.
A mirror-type children's sitting posture monitoring robot is designed to detect the child's sitting posture through the camera and visual recognition module, and use the angle adjustment mechanism to imitate the child's sitting posture, adjust the tilt angle of the robot to correct the child's sitting posture.
It realizes sitting posture correction without forcibly restraining the child's posture, avoiding negative impacts on physical health, with many monitoring points, many postures and high accuracy.
Smart Images

Figure CN118990542B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of children's sitting posture correction equipment, and in particular to a mirror-type children's sitting posture monitoring robot. Background Art
[0002] Parents are paying more and more attention to their children's growth and development. However, since children need to complete homework, study and reading tasks in and out of class, and do not have the habit of maintaining a good sitting posture, children and adolescents have a high incidence of myopia, uneven shoulders, hunchbacks and other problems.
[0003] Passive posture correction products such as Beibeijia, anti-hunchback support, and anti-hunchback beautiful buttocks cushion will forcibly restrain the child's body posture, causing problems such as bone deformation. In severe cases, problems such as thoracic kyphosis and lumbar protrusion may occur. Active posture correction products such as Dali Desk Lamp and Jiaozi Elf have too few monitoring points, low monitoring accuracy, and frequent false alarms, which will eventually lead to product abandonment.
[0004] To this end, a mirror-type child sitting posture monitoring robot is designed to provide another technical solution to the above technical problems. Summary of the invention
[0005] Based on this, it is necessary to provide a mirror-type child sitting posture monitoring robot to solve the technical problems raised in the above-mentioned background technology.
[0006] In order to solve the above technical problems, the present invention adopts the following technical solutions:
[0007] A mirror-type child sitting posture monitoring robot comprises a fixed shell, a fixed body is fixed inside the top of the fixed shell, a camera is fixed inside the bottom of one side of the fixed body, a base is fixed at the bottom of the fixed shell, an angle adjustment mechanism for adjusting the center of gravity is installed on the top of the base, the angle adjustment mechanism comprises a plane angle adjustment rotating motor, a support platform, an inclination angle adjustment rotating motor, a U-shaped frame and a counterweight, the bottom of the plane angle adjustment rotating motor is fixed to the base, the output end of the plane angle adjustment rotating motor is fixed with a support platform, the top of the support platform is fixed with a inclination angle adjustment rotating motor, the output end of the inclination angle adjustment rotating motor is bolted to the U-shaped frame, and the top of the U-shaped frame is fixed with a counterweight.
[0008] As a preferred implementation of the mirror-type child sitting posture monitoring robot provided by the present invention, the fixed shell, the fixed body, the camera and the base are assembled in an outer shape of a tumbler.
[0009] As a preferred implementation of the mirror-type child sitting posture monitoring robot provided by the present invention, a spring is fixed to the bottom of the base, and a positioning suction cup is fixed to the bottom of the spring.
[0010] As a preferred implementation of the mirror-type child sitting posture monitoring robot provided by the present invention, the center of the counterweight block coincides with the center of the base.
[0011] As a preferred embodiment of the mirror-type child sitting posture monitoring robot provided by the present invention, a first rotation angle sensor is fixed to the output end of the plane angle adjustment rotating motor, and a second rotation angle sensor is fixed to the output end of the tilt angle adjustment rotating motor.
[0012] As a preferred embodiment of the mirror-type child sitting posture monitoring robot provided by the present invention, a visual recognition module is arranged inside the fixed body, which is used to calculate the current child sitting posture inclination angle through the child sitting posture photographed by the camera;
[0013] The visual recognition module includes a sitting posture comparison module and an adjustment control module;
[0014] The sitting posture comparison module is used to determine whether the current child sitting posture is correct through the current child sitting posture picture;
[0015] The adjustment control module is used to select the tilt angle adjustment rotary motor to work alone or the plane angle adjustment rotary motor and the tilt angle adjustment rotary motor to work simultaneously according to the signal given by the feature comparison module to adjust the angle of the robot.
[0016] As a preferred embodiment of the mirror-type child sitting posture monitoring robot provided by the present invention, the child sitting posture determination module includes a data collection module, a data processing module, a feature extraction module and a feature comparison module;
[0017] The data collection module is used to collect children's sitting posture image data, including children's correct sitting posture pictures, incorrect sitting posture pictures and transition states between different sitting postures;
[0018] The data processing module is used to process the sitting posture pictures collected by the data collection module;
[0019] The feature extraction module is used to extract key points of the child's sitting posture in the picture processed by the data processing module;
[0020] The feature comparison module takes the coordinate points of the left mouth corner MOUTH_LEFT and the left ear LEFT_EAR based on the feature points of each part extracted by the feature extraction module, calculates the vector l eft_vex pointing from the left ear to the left mouth corner, takes the coordinate points of the right mouth corner MOUTH_RIGHT and the right ear RIGHT_EAR, calculates the vector right_vex pointing from the right ear to the right mouth corner, adds the two vectors to obtain the vex vector, takes the coordinates of the left shoulder LEFT_SHOULDER and the right shoulder RIGHT_SHOULDER, and calculates the vector shoulder_vew pointing from the left shoulder to the right shoulder.
[0021] As a preferred embodiment of the mirror-type child sitting posture monitoring robot provided by the present invention, the angle between vex and shoulder_vex is calculated, and the angle is compared with the angle within the normal physiological range to determine whether the sitting posture is bent forward and backward;
[0022] Calculate the angle between shoulder_vew and the horizontal vector, and compare this angle with the angle within the normal physiological range to determine whether scoliosis occurs;
[0023] The expression is as follows:
[0024]
[0025] As a preferred embodiment of the mirror-type child sitting posture monitoring robot provided by the present invention, the adjustment control module is used to select the tilt angle adjustment rotary motor to work alone or the plane angle adjustment rotary motor and the tilt angle adjustment rotary motor to work simultaneously according to the signal given by the feature comparison module to adjust the angle of the robot. The specific steps are as follows:
[0026] A. The adjustment control module is used to judge whether the plane angle adjustment rotary motor and the tilt angle adjustment rotary motor need to work, and the rotation angles of the plane angle adjustment rotary motor and the tilt angle adjustment rotary motor according to the signal given by the feature comparison module, and when the plane angle adjustment rotary motor and the tilt angle adjustment rotary motor need to work at the same time, enter step B, and when the plane angle adjustment rotary motor does not need to work, enter step C;
[0027] B. Start the plane angle adjustment rotary motor to adjust the horizontal angle of the support platform to the corresponding value, and start the tilt angle adjustment rotary motor to adjust the tilt angle of the U-shaped frame and the counterweight to the corresponding value, and then go to step D;
[0028] C. Start the tilt angle adjustment rotating motor, adjust the tilt angle of the U-shaped frame and the counterweight to the corresponding value, and go to step D;
[0029] D. Determine whether the robot's tilt angle is correct. If it is correct, proceed to step E. If not, proceed to step A.
[0030] E. Wait for new signal from the adjustment control module and continue to repeat steps A to D.
[0031] It can be seen without a doubt that the above-mentioned technical solution of the present application can definitely solve the technical problem to be solved by the present application.
[0032] At the same time, through the above technical solutions, the present invention has at least the following beneficial effects:
[0033] The mirror-type child sitting posture monitoring robot provided by the present invention can make the robot imitate the child's sitting posture by inclining the center of gravity column in the angle adjustment mechanism according to the child's sitting posture detected by the visual recognition module when the robot is in use, so that the child can correct his or her sitting posture after seeing the robot, thereby making it possible to correct the child's sitting posture without forcibly restraining the child's posture, and thus will not have a negative impact on the child's health. The robot has many monitoring points, can identify many postures, and has high accuracy.
[0034] By cooperating with the plane angle adjustment rotating motor, the tilt angle adjustment rotating motor and the counterweight, the tilt angle adjustment rotating motor can work alone to allow the counterweight to rotate approximately forty-five degrees opposite to the child. By working simultaneously with the plane angle adjustment rotating motor and the vertical angle adjustment rotating motor, the counterweight can be rotated to any position of approximately forty-five degrees above the desktop, thereby adjusting the angle of the monitoring robot on the desktop. BRIEF DESCRIPTION OF THE DRAWINGS
[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0036] Figure 1 It is a schematic diagram of the overall structure of the present invention;
[0037] Figure 2 It is a structural schematic diagram of the base of the present invention;
[0038] Figure 3 It is a structural schematic diagram of the angle adjustment mechanism of the present invention;
[0039] Figure 4 This is a detection flow chart of Embodiment 2 of the present invention;
[0040] Figure 5 This is a flowchart of angle determination and adjustment according to Embodiment 2 of the present invention;
[0041] Figure 6 Schematic diagram of visual recognition points according to the second embodiment of the present invention.
[0042] In the figure: 1. fixed shell; 2. fixed body; 3. camera; 4. base; 5. plane angle adjustment rotating motor; 6. support platform; 7. tilt angle adjustment rotating motor; 8. U-shaped frame; 9. counterweight; 10. spring; 11. positioning suction cup. DETAILED DESCRIPTION
[0043] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.
[0044] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings.
[0045] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features and technical solutions in the embodiments may be combined with each other.
[0046] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.
[0047] Embodiment 1
[0048] Reference Figure 1-Figure 3 A mirror-type child sitting posture monitoring robot comprises a fixed shell 1, a fixed body 2 is fixed inside the top of the fixed shell 1, a camera 3 is fixed inside the bottom of one side of the fixed body 2, so that the sitting posture of the child can be photographed and identified through the operation of the camera 3, a base 4 is fixed at the bottom of the fixed shell 1, so that the base 4 can be fixed to the fixed shell 1 by screws or bonding, and the bottom end of the base 4 is filled with materials, so that the base 4 will not tilt when there is no external force, and the fixed shell 1, the fixed body 2, the camera 3 and the base 4 are assembled in the shape of a tumbler, so that even if the monitoring robot is tilted during use, it can be reset after the external force disappears;
[0049] In this embodiment, the camera 3 is composed of two existing camera devices, so that the viewing angle of the monitored child can be more clearly and completely identified.
[0050] A spring 10 is fixed to the bottom of the base 4, and a positioning suction cup 11 is fixed to the bottom of the spring 10, so that the positioning suction cup 11 can be installed on the desktop through adsorption, so that the position of the base 4 can be positioned, and the center of gravity adjustment of the base 4 caused by external force drives the spring 10 to bend, so that after the external force disappears, the monitoring robot can be reset by the reset of the spring 10 after bending.
[0051] An angle adjustment mechanism for adjusting the center of gravity is installed on the top of the base 4, and the angle adjustment mechanism includes a plane angle adjustment rotating motor 5, a support platform 6, a tilt angle adjustment rotating motor 7, a U-shaped frame 8 and a counterweight 9. The bottom of the plane angle adjustment rotating motor 5 is fixed to the base 4, and the output end of the plane angle adjustment rotating motor 5 is fixed with the support platform 6, so that the plane angle adjustment rotating motor 5 drives the support platform 6 to rotate in the plane, and the tilt angle adjustment rotating motor 7 is fixed inside the top of the support platform 6. The output end of the tilt angle adjustment rotating motor 7 is bolted with the U-shaped frame 8, so that the rotation of the tilt angle adjustment rotating motor 7 can drive the U-shaped frame 8 to rotate, and then the U-shaped frame 8 can be adjusted with respect to the desktop, and a counterweight 9 is fixed on the top of the U-shaped frame 8, and the center of the counterweight 9 coincides with the center of the base 4, so that when the counterweight 9 follows the rotation of the U-shaped frame 8 for adjustment, the angle between the monitoring robot and the desktop is adjusted;
[0052] In this embodiment, when the U-shaped frame 8 and the counterweight block 9 are viewed from a top view, their combined shape is a cross shape. In other embodiments, the combined shape can be changed according to specific requirements such as the center of gravity.
[0053] In this embodiment, the tilt angle adjustment rotary motor 7 is a dual-axis motor.
[0054] A first rotation angle sensor is fixed to the output end of the plane angle adjustment rotary motor 5, and a second rotation angle sensor is fixed to the output end of the tilt angle adjustment rotary motor 7. The first rotation angle sensor and the second rotation angle sensor are both of the same type.
[0055] SV01A103AEA01R00 can detect the rotation angles of the plane angle adjustment rotary motor 5 and the tilt angle adjustment rotary motor 7 during operation through the first rotation angle sensor and the second rotation angle sensor, thereby being able to accurately control the tilt angle of the robot.
[0056] The use process of the mirror-type child sitting posture monitoring robot provided by the present invention is as follows: when in use, the monitoring robot is positioned by adsorbing the bottom of the positioning suction cup 11 to the desktop, and then when the camera 3 detects that the sitting posture of the child is tilted, according to the tilt angle of the child's sitting posture, the tilt angle adjustment rotary motor 7 is selected to work alone or the plane angle adjustment rotary motor 5 and the tilt angle adjustment rotary motor 7 are selected to work simultaneously;
[0057] When it is detected that the sitting posture of the child is tilted to the left or right, when the tilt angle adjustment rotary motor 7 is selected to work alone, the U-shaped frame 8 is driven to rotate by the tilt angle adjustment rotary motor 7, so that the rotation of the U-shaped frame 8 drives the counterweight block 9 to rotate, and then the angle between the counterweight block 9 and the desktop is adjusted, so as to adjust the center of gravity of the monitoring robot, so as to adjust the tilt angle of the robot, and the second rotation angle sensor on the tilt angle adjustment rotary motor 7 detects the rotation angle of the support platform 6 when it is working, so that the robot can rotate about plus or minus forty-five degrees opposite to the child;
[0058] When it is monitored that the child's sitting posture is tilted to the front or back, the left front or the right front or the left back or the right back, the plane angle adjustment rotating motor 5 is first rotated to drive the tilt angle adjustment rotating motor 7 to rotate in the plane. When it rotates to the specified angle, the tilt angle adjustment rotating motor 7 drives the counterweight 9 to rotate through the U-shaped frame 8 to tilt the robot to the corresponding angle. The robot can be rotated to any position of plus or minus about forty-five degrees above the desktop through the rotation adjustment of the counterweight 9, thereby adjusting the angle of the monitoring robot on the desktop.
[0059] Embodiment 2
[0060] Reference Figure 4-Figure 6 , a mirror-type child sitting posture monitoring robot, a visual recognition module is arranged inside the fixed body 2, which is used to calculate the current child sitting posture inclination angle through the child sitting posture photographed by the camera 3;
[0061] The visual recognition module includes a sitting posture comparison module and an adjustment control module;
[0062] The sitting posture comparison module is used to determine whether the current child's sitting posture is correct based on the current child's sitting posture picture;
[0063] The sitting posture comparison module includes a child sitting posture determination module and an output instruction module;
[0064] The child sitting posture determination module is used to extract the child sitting posture module feature points captured by the camera 3 through the trained model, and determine whether the child sitting posture is correct based on the comparison of the feature points;
[0065] The child sitting posture determination module includes a data collection module, a data processing module, a feature extraction module and a feature comparison module;
[0066] The data collection module is used to collect children's sitting posture image data, including children's correct sitting posture pictures, incorrect sitting posture pictures (bent over sitting posture, hunched over sitting posture, etc.) and transition states between different sitting postures. Specifically, it can be shot with a professional camera or cropped after the overall recording by a high-definition video recording device, while ensuring the quality and clarity of the pictures or videos;
[0067] The data processing module is used to process the sitting posture pictures collected by the data collection module, so as to facilitate feature extraction of the sitting posture pictures;
[0068] The data processing module includes a noise processing module and an image cropping module;
[0069] The noise processing module is used to perform weighted averaging on each pixel in the image according to the weight of the Gaussian function through Gaussian filtering to smooth the image and reduce noise. The specific steps are as follows:
[0070] When the OpenCV library in Python implements Gaussian filtering to remove noise, it is as follows:
[0071] import cv2
[0072] #Read the original image
[0073] image=cv2.imread('noisy_image.jpg',cv2.IMREAD_GRAYSCALE)#If the image is already a grayscale image, you can omit I MREAD_GRAYSCALE
[0074] # Apply Gaussian filter
[0075] smoothed_image = cv2.GaussianB l ur(image,(5,5),0) #(5,5) is the size of the filter, 0 is the standard deviation of the Gaussian distribution
[0076] # Display the filtered image
[0077] cv2.imshow('Smoothed Image',smoothed_image)
[0078] # Wait for keyboard input
[0079] cv2.waitKey(0)
[0080] #Close the window
[0081] cv2.destroyAllWidows()
[0082] In the above code, the cv2.GaussianBl ur function accepts three parameters:
[0083] The first argument is the image to be filtered.
[0084] The second parameter is a tuple indicating the size of the filter. Usually, the filter size is an odd number, such as 5x5, 7x7, etc.
[0085] The third parameter is the standard deviation of the Gaussian distribution. The larger the standard deviation, the larger the filter's range of influence and the better the noise removal effect, but it may also smooth out useful details in the image. The value of the standard deviation usually needs to be adjusted according to the specific noise level and image content.
[0086] The image cropping module is used to remove the redundant parts of the image, thereby retaining the child's sitting area. The specific steps are as follows:
[0087] #Crop the image
[0088] cropped_image=image[y1:y2,x1:x2]#crop the image according to the coordinates of the upper left corner and the lower right corner
[0089] # Display results
[0090] cv2.imshow('Cropped Image',cropped_image)
[0091] cv2.waitKey(0)
[0092] cv2.destroyAllWidows()
[0093] The feature extraction module is used to extract the key points of the child's sitting posture in the picture processed by the data processing module;
[0094] The feature extraction module includes a target detection module, a posture punctuation module and a feature extraction module;
[0095] The object detection module is used to detect the location of children in the image using the YOLO (You On ly Look Once) algorithm;
[0096] The pose punctuation module is used to detect the position of the child's head, torso, neck, shoulder, elbow, wrist, hip, knee and ankle through the OpenPose algorithm;
[0097] The feature extraction module is used to calculate the position detected by the posture punctuation module and the body's tilt angle. The steps are as follows:
[0098] The Sobel edge detection algorithm can detect the edges in the image to determine the position of the head;
[0099] The trunk key points and thigh key points can form a vector, and the angle between these two vectors can be used to calculate the inclination angle of the child's trunk.
[0100] The color distribution and changes in the image are analyzed by color histogram to determine the tilt angle of the body, so that the tilt angle of the body can be estimated by analyzing the color distribution of different parts in the image;
[0101] Therefore, according to the head position: observe whether the child's head remains upright, whether it leans forward or backward; shoulder position: check whether the shoulders are parallel to the ground, whether the shoulders are relaxed and not hunched; spinal curve: assess whether the child's spine maintains a natural S-shaped curve, whether there is excessive bending or undercorrection; arms and wrists: observe whether the child's arms and wrists are relaxed when sitting, whether they are excessively bent or stretched; leg position: check whether the legs are relaxed and straight, whether the feet are flat on the ground, and whether the knees are relaxed; body balance: assess whether the child can maintain a stable posture when sitting, whether there is swaying or needing support; breathing status: observe whether the child can breathe naturally when sitting, and whether breathing is affected by improper posture; duration: record the time the child can maintain the correct sitting posture to assess its persistence; concentration: observe whether the child can concentrate when sitting, and whether there is distracting behavior; adaptability: assess the child's sitting performance in different environments or different activities to observe his or her adaptability.
[0102] Based on the feature points of each part extracted by the feature extraction module, the feature comparison module takes the coordinate points of the left corner of the mouth MOUTH_LEFT and the left ear LEFT_EAR, and calculates the vector l eft_vex pointing from the left ear to the left corner of the mouth; similarly, it takes the coordinate points of the right corner of the mouth MOUTH_RIGHT and the right ear RI GHT_EAR, calculates the vector ri ght_vex pointing from the right ear to the right corner of the mouth, and adds the two vectors to obtain the vex vector.
[0103] Take the coordinates of the left shoulder LEFT_SHOULDER and the right shoulder RIGHT_SHOULDER, and calculate the vector shoulder_vew pointing from the left shoulder to the right shoulder.
[0104] According to the angle calculation formula between two vectors:
[0105]
[0106] The angle between vex and shoulder_vex can be calculated. By comparing this angle with the angle within the normal physiological range, it can be determined whether the sitting posture is bent forward or backward.
[0107] Similarly, by calculating the angle between shoulder_vew and the horizontal vector, it can be determined whether scooting occurs.
[0108] The adjustment control module is used to select the tilt angle adjustment rotary motor 7 to work alone or the plane angle adjustment rotary motor 5 and the tilt angle adjustment rotary motor 7 to work simultaneously according to the signal given by the feature comparison module, so as to adjust the angle of the robot, thereby guiding children with poor sitting posture to gradually adjust their sitting posture in a humanized way, and finally helping children to develop correct sitting habits; the specific steps are as follows:
[0109] A. The adjustment control module is used to judge whether the plane angle adjustment rotary motor 5 and the tilt angle adjustment rotary motor 7 need to work, and the rotation angles of the plane angle adjustment rotary motor 5 and the tilt angle adjustment rotary motor 7 according to the signal given by the feature comparison module, and when the plane angle adjustment rotary motor 5 and the tilt angle adjustment rotary motor 7 need to work at the same time, it goes to step B, and when the plane angle adjustment rotary motor 5 does not need to work, it goes to step C;
[0110] B. Start the plane angle adjustment rotary motor 5 to adjust the horizontal angle of the support platform 6 to the corresponding value, and start the tilt angle adjustment rotary motor 7 to adjust the tilt angle of the U-shaped frame 8 and the counterweight block 9 to the corresponding value, and then proceed to step D;
[0111] C. Start the tilt angle adjustment rotating motor 7, adjust the tilt angles of the U-shaped frame 8 and the counterweight 9 to corresponding values, and proceed to step D;
[0112] D. Determine whether the robot's tilt angle is correct. If it is correct, proceed to step E. If not, proceed to step A.
[0113] E. Wait for new signal from the adjustment control module and continue to repeat steps A to D.
[0114] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the present invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.
Claims
1. A mirror-type child sitting posture monitoring robot, characterized in that: The invention comprises a fixed shell (1), wherein a fixed body (2) is fixed inside the top of the fixed shell (1), a camera (3) is fixed inside the bottom of one side of the fixed body (2), a base (4) is fixed at the bottom of the fixed shell (1), an angle adjustment mechanism for adjusting the center of gravity is installed on the top of the base (4), the angle adjustment mechanism comprises a plane angle adjustment rotating motor (5), a support platform (6), an inclination angle adjustment rotating motor (7), a U-shaped frame (8) and a counterweight (9), the bottom of the plane angle adjustment rotating motor (5) is fixed to the base (4), the output end of the plane angle adjustment rotating motor (5) is fixed with a support platform (6), the top of the support platform (6) is fixed with an inclination angle adjustment rotating motor (7), the output end of the inclination angle adjustment rotating motor (7) is bolted to a U-shaped frame (8), and the top of the U-shaped frame (8) is fixed with a counterweight (9); A spring (10) is fixed to the bottom of the base (4), and a positioning suction cup (11) is fixed to the bottom of the spring (10); The center of the counterweight block (9) coincides with the center of the base (4); A first rotation angle sensor is fixed to the output end of the plane angle adjustment rotary motor (5), and a second rotation angle sensor is fixed to the output end of the tilt angle adjustment rotary motor (7).
2. The mirror-type child sitting posture monitoring robot according to claim 1, characterized in that: The fixed shell (1), the fixed body (2), the camera (3) and the base (4) have an appearance of a tumbler after being assembled.
3. The mirror-type child sitting posture monitoring robot according to claim 1, characterized in that: A visual recognition module is provided inside the fixed body (2) for calculating the current inclination angle of the child's sitting posture through the child's sitting posture photographed by the camera (3); The visual recognition module includes a sitting posture comparison module and an adjustment control module; The sitting posture comparison module is used to determine whether the current child sitting posture is correct through the current child sitting posture picture; The adjustment control module is used to select, based on the signal given by the feature comparison module, to allow the tilt angle adjustment rotary motor (7) to work alone or the plane angle adjustment rotary motor (5) and the tilt angle adjustment rotary motor (7) to work simultaneously, thereby adjusting the angle of the robot.
4. The mirror-type child sitting posture monitoring robot according to claim 3, characterized in that: The sitting posture comparison module includes a data collection module, a data processing module, a feature extraction module and a feature comparison module; The data collection module is used to collect children's sitting posture image data, including children's correct sitting posture pictures, incorrect sitting posture pictures and transition states between different sitting postures; The data processing module is used to process the sitting posture pictures collected by the data collection module; The feature extraction module is used to extract key points of the child's sitting posture in the picture processed by the data processing module; The feature comparison module takes the coordinate points of the left mouth corner MOUTH_LEFT and the left ear LEFT_EAR based on the feature points of each part extracted by the feature extraction module, calculates the vector left_vex pointing from the left ear to the left mouth corner, takes the coordinate points of the right mouth corner MOUTH_RIGHT and the right ear RIGHT_EAR, calculates the vector right_vex pointing from the right ear to the right mouth corner, adds the two vectors to obtain the vex vector, takes the coordinates of the left shoulder LEFT_SHOULDER and the right shoulder RIGHT_SHOULDER, and calculates the vector shoulder_vew pointing from the left shoulder to the right shoulder.
5. The mirror-type child sitting posture monitoring robot according to claim 4, characterized in that: Calculate the angle between vex and shoulder_vex, and compare this angle with the angle within the normal physiological range to determine whether the sitting posture is bent forward or backward; Calculate the angle between shoulder_vew and the horizontal vector, and compare this angle with the angle within the normal physiological range to determine whether scoliosis occurs; The expression is as follows:
6. The mirror-type child sitting posture monitoring robot according to claim 3, characterized in that: The adjustment control module is used to select, according to the signal given by the feature comparison module, to let the tilt angle adjustment rotary motor (7) work alone or the plane angle adjustment rotary motor (5) and the tilt angle adjustment rotary motor (7) work simultaneously, so as to adjust the angle of the robot. The specific steps are as follows: A. The adjustment control module is used to judge whether the plane angle adjustment rotating motor (5) and the tilt angle adjustment rotating motor (7) need to work, and the rotation angles of the plane angle adjustment rotating motor (5) and the tilt angle adjustment rotating motor (7) according to the signal given by the feature comparison module. When the plane angle adjustment rotating motor (5) and the tilt angle adjustment rotating motor (7) need to work at the same time, the process proceeds to step B; when the plane angle adjustment rotating motor (5) does not need to work, the process proceeds to step C; B. Start the plane angle adjustment rotary motor (5) to adjust the horizontal angle of the support platform (6) to a corresponding value, and start the tilt angle adjustment rotary motor (7) to adjust the tilt angles of the U-shaped frame (8) and the counterweight (9) to corresponding values, and proceed to step D; C. Start the tilt angle adjustment rotating motor (7), adjust the tilt angles of the U-shaped frame (8) and the counterweight (9) to corresponding values, and proceed to step D; D. Determine whether the robot's tilt angle is correct. If it is correct, proceed to step E. If not, proceed to step A. E. Wait for new signal from the adjustment control module and continue to repeat steps A to D.
Citation Information
Patent Citations
Simulation interaction device and method for sitting postures of children
CN108562272A
Abnormal behavior identification method and system for security and protection monitoring
CN118155126A